|Home | About | Journals | Submit | Contact Us | Français|
The binding of high density lipoprotein (HDL) to scavenger receptor BI (SR-BI) is responsible for whole-body cholesterol disposal via reverse cholesterol transport. The extracellular domain of SR-BI is required for HDL binding and selective uptake of HDL-cholesterol. We identified six highly hydrophobic regions in this domain that may be important for receptor activity and performed site-directed mutagenesis to investigate the importance of these regions in SR-BI-mediated cholesterol transport. Non-conservative mutation of the regions encompassing V67, L140/L142, V164 or V221 reduced hydrophobicity and impaired the ability of SR-BI to bind HDL, mediate selective uptake of HDL-cholesterol, promote cholesterol efflux, and enlarge the cholesterol oxidase-sensitive pool of membrane free cholesterol. In contrast, conservative mutations at V67, V164 or V221 did not affect the hydrophobicity or these cholesterol transport activities. We conclude that the hydrophobicity of N-terminal extracellular regions of SR-BI is critical for cholesterol transport, possibly by mediating receptor-ligand and/or receptor-membrane interactions.
The inverse correlation between the risk for developing coronary artery disease and plasma concentrations of high density lipoprotein (HDL)1 [1, 2] has been attributed to the strong athero-protective effects of HDL that include inhibition of low density lipoprotein oxidation [3, 4] and oxidative damage , promotion of endothelial nitric oxide production [6, 7] and vascular reactivity and integrity , inhibition of platelet aggregation and coagulation [9, 10] and prevention of thrombosis . However, the primary athero-protective role of HDL stems from its ability to promote the disposal of peripheral cholesterol at the liver via a process termed reverse cholesterol transport .
The final step of reverse cholesterol transport involves the movement of cholesterol from HDL to the liver for catabolism. The selective transfer of cholesteryl ester (CE) from HDL to cells is mediated by scavenger receptor class B type I (SR-BI) , an 82-kDa glycosylated cell surface receptor  highly expressed in the liver and steroidogenic tissues [15–17]. SR-BI (509 amino acids) consists of a large extracellular domain (403 amino acids) anchored by two transmembrane domains and two short cytoplasmic tails . Transgenic overexpression [19–21] or hepatic adenoviral infection [22, 23] of SR-BI decreased HDL plasma cholesterol levels and increased cholesterol catabolism and excretion. On the other hand, a 50% reduction in SR-BI expression  or full disruption of the SR-BI gene in mice increased plasma HDL-cholesterol levels and reduced neutral lipid stores in the adrenal gland and ovary [24, 25]. Thus, SR-BI is the most physiologically relevant HDL receptor.
SR-BI-mediated selective uptake of HDL-CE is a two-step process: (i) HDL must bind to the extracellular domain of SR-BI and (ii) CE is transferred from HDL to the plasma membrane by a non-endocytic mechanism, without holoparticle uptake or degradation of apolipoproteins [26–28]. The critical nature of the extracellular domain of SR-BI in CE transfer has been demonstrated through the use of chimeric receptors [29–31] and insertion of epitope tags into various regions of the extracellular domain of SR-BI . Moreover, antibodies to the extracellular domain blocked HDL-CE-selective uptake and the delivery of HDL-CE to the steroidogenic pathway in cultured adrenocortical cells . In fact, a set of distinct SR-BI-mediated activities appears to be inherent to the extracellular domain, including free cholesterol (FC) efflux and influx, as well as the ability to increase cellular FC mass and enhance sensitivity of membrane FC to exogenous cholesterol oxidase .
Our detailed analyses also reveal the presence of evolutionarily conserved sequences with high hydrophobicity within the extracellular domain of SR-BI. We hypothesized that these hydrophobic regions may play a role in mediating the cholesterol transport functions of SR-BI. To test this hypothesis, we used site-directed mutagenesis to generate point mutations that would reduce overall hydrophobicity of the particular regions: V67N, L140Q/L142Q, V164N, V221N, L359Q, and L411Q. We then correlated the changes in hydrophobicity to the effects on HDL binding, selective uptake of HDL-CE and other functions of SR-BI. In addition, we created a second set of point mutations that maintained the overall hydrophobicity of the selected regions (V67L, L140V/L142V, V164L, V221L, L359V, and L411V) to test whether the changes in SR-BI function were due to changes in hydrophobicity or changes in amino acid identity.
The following antibodies were used: polyclonal anti-SR-BI specific for the C-terminal or the extracellular domain (Novus Biologicals, Inc., Littleton, CO); anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Millipore, Billerica, MA); peroxidase-conjugated goat anti-rabbit secondary IgG (Jackson ImmunoResearch Laboratories, West Grove, PA). Human HDL (1.063–1.21 g/mL) was purchased from Biomedical Technologies, Inc. [125I]Iodine was from Perkin-Elmer, while [3H]cholesterol and [3H]cholesteryl oleoyl ether (COE) were from GE Healthcare (Piscataway, NJ). Cholesterol oxidase (Nocardia erythropolis) was from MP Biomedicals, LLC. Acyl-CoA:cholesterol acyltransferase (ACAT) inhibitor (Sandoz 58-035), perfluoro-octanoic acid (PFO) cholesterol, 4-cholesten-3-one and cholesteryl oleate standards were purchased from Sigma. All other reagents were of analytical grade.
Site-directed mutations of V67, V164, V221 and L359 were introduced into wild-type murine SR-BI (pSG5(SR-BI))  using the QuikChange Site-Directed Mutagenesis kit (Stratagene) according to manufacturer’s protocols. Oligonucleotide primers were purchased from Integrated DNA Technologies. The following primers were used for mutagenesis: V67N, 5′-CTA CTT GTC TAA CTA CTT CTT CG -3′ and 5′ – CGA AGA AGT AGT TAG ACA AGT AG -3′; V164N, 5′-CCT TGG CGC TGA ACA CCA TGG GCC -3′ and 5′-GGC CCA TGG TGT TCA GCG CCA AGG -3′; V221N, 5′-GGG GTC TTC ACT AAC TTC ACG GGC GTC -3′ and 5′-GAC GCC CGT GAA GTT AGT GAA GAC CCC -3′; L359Q, 5′-CAG AAG CTG TTC AAG GTC TGA ACC C -3′ and 5′-GGG TTC AGA CCT TGA ACA GCT TCT G -3′. All plasmids were purified using endotoxin-free Qiagen Maxi-Prep kits and sequenced through the coding region to ensure the presence of the correct mutation and the absence of undesired mutations generated during the amplification steps. DNA sequencing was performed on an ABI 3100 at the Protein and Nucleic Acid Facility at the Medical College of Wisconsin. The L140Q/L142Q-, L411Q-, V67L-, L140V/L142V-, V164L-, V221L-, L359V- and L411V-SR-BI mutant receptors were made and sequenced by TOP Gene Technologies (Pointe-Claire, Quebec, Canada).
COS-7 cells were maintained in DMEM (Invitrogen), 10% calf serum (Invitrogen), 2 mM L-glutamine, 50 units/ml penicillin, 50 μg/ml streptomycin and 1 mM sodium pyruvate, and transfected as previously described . The following day, fresh medium was added and the cells were assayed 48 h post-transfection, with the exception of efflux assays where cells were assayed 72 h post-transfection.
Transiently-transfected cells expressing murine wild-type or mutant SR-BI were washed twice in PBS, pH 7.4, and lysed with 300 μl of 1% NP-40 cell lysis buffer  containing protease inhibitors (1 μg/ml pepstatin, 0.2 mM phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 10 μg/ml aprotinin). Protein concentrations were determined by the Lowry method .
HDL was labeled with dilactitol tyramine and non-hydrolyzable [3H]COE as described . Average specific activities were 8.3 dpm/ng protein for 3H and 330.4 dpm/ng protein for 125I.
COS-7 cells were transiently transfected with pSG5 or vectors encoding wild-type and mutant SR-BI cDNAs. Cell association of [125I]HDL and selective uptake assays of non-hydrolyzable [3H]COE were performed as previously described . Data represent the average of four separate experiments performed in triplicate. Data were calculated as ng HDL/mg cell protein or ng HDL-COE/mg cell protein for binding and selective uptake experiments, respectively. Empty vector values were subtracted from all wild-type and mutant values, and data were normalized to wild-type SR-BI. Statistical comparisons were calculated by one-way ANOVA with Bonferroni post-tests for all groups.
Assays were performed as previously described , unless otherwise specified. Data represent the average of four separate experiments, each performed in quadruplicate. Statistical comparisons were calculated by one-way ANOVA with Bonferroni post-tests for all groups.
Transiently-transfected COS-7 cells were re-plated 24 hours post-transfection onto glass coverslips in 12-well plates. After 24 hours, the media were removed, cells were washed with PBS and fixed with 4% paraformaldehyde/PBS for 10 minutes. Cells were then stained as described by Peng et al.  using a primary antibody directed against the extracellular domain of SR-BI (1:200) and an Alexa 488-conjugated goat anti-rabbit IgG secondary antibody (Molecular Probes, Inc., 1:1000). Cells were examined on a Leica TCS SP2 Laser Scanning confocal microscope imaging system.
Transiently-transfected cells expressing wild-type or mutant SR-BI receptors were washed twice with PBS, and lysed with 1% NP-40 cell lysis buffer containing the above-mentioned protease inhibitors. Equal amounts of total cellular proteins were separated on 8% SDS-PAGE gels, transferred to nitrocellulose membranes (GE Healthcare, Piscataway, NJ) and detected using an antibody specific for the C-terminal domain of SR-BI followed by horseradish peroxidase-conjugated anti-rabbit secondary antibody. Antigen-antibody complexes were visualized with SuperSignal West Pico reagent (Pierce).
Transiently-transfected COS-7 cells expressing wild-type or mutant SR-BI receptors were washed twice with PBS and incubated with 1 mg/mL EZ-Link Sulfo-NHS-LC-biotin (Pierce) in PBS for 30 min at 4 °C in the dark. Cells were lysed with 1% NP-40 cell lysis buffer containing the above-mentioned protease inhibitors. A portion of the cell lysate was incubated with streptavidin agarose (Pierce) for 1 hour at room temperature. The beads were pelleted, washed 4 times in cell lysis buffer and resuspended in sample treatment buffer. Biotinylated SR-BI was detected by immunoblot analysis as described above.
PFO-PAGE was performed as described , but with the following modifications. Forty-eight hours post-transfection, COS7 cells expressing wild-type or mutant SR-BI receptors were washed twice with PBS and lysed in PBS containing the aforementioned protease inhibitors. Prior to electrophoresis, lysates were sonicated using a Fisher Scientific 550 Sonic Dismembrator on ice five times at 30% power (power level 3) for 3 s each, with a 1-min cooling period between sonications. Equal amounts of total cellular proteins were solubilized in sample treatment buffer (5% PFO, 100 mM Tris, 20% glycerol, 0.005% bromophenol blue). Samples were rotated for 30 min at room temperature and separated on 5% polyacrylamide gels (lacking SDS) in running buffer (25 mM Tris, 192 mM glycine, 0.5% PFO, final pH 8.5). Proteins were transferred to nitrocellulose membranes and detected by immunoblot analysis as described above.
The traditional depiction of SR-BI with a “horseshoe-like” topology is based upon hydropathy analysis  of the predicted amino acid sequence and reveals two membrane-spanning regions. These N- and C-terminal domains display a hydrophobicity index much greater than the typical threshold value and are long enough to span the membrane bilayer [39, 40]. However, the hydropathy plot (Figure 1) also revealed that the extracellular domain of SR-BI contains other evolutionarily conserved sequences with a high hydrophobicity index (Figure 2).
In order to examine the functional role of these hydrophobic regions, we created a panel of strategic point mutations that greatly reduced the overall hydrophobicity of the given region . In this panel, leucine (Leu, L) or valine (Val, V) residues were mutated to either glutamine (Gln, Q) or asparagine (Asn, N), respectively, to generate non-conservative substitutions. We also developed a second group of point mutations designed to maintain the overall hydrophobicity of the given region. In this group, conservative substitutions were created by mutating Leu residues to Val residues, and Val residues to Leu residues. Table I outlines the amino acids chosen for mutagenesis and the effect of these mutations on the overall hydrophobicity of that region. We chose to design the L140Q/L142Q double point mutation since mutation of both Leu residues resulted in a greater decrease in hydropathy index than mutating either one individually; therefore, our second group of point mutations also contained a double point mutation (L140V/L142V). Further, although residues at positions 164 and 359 are not identical among species, the hydrophobicity of the region remains high among species (hydrophobicity indices range from 1.69–2.17 or 1.77–2.17 for windows containing residue 164 or 359, respectively).
Plasmids encoding wild-type and mutant SR-BI were transiently-transfected into COS-7 cells and examined for expression by fluorescent microscopy. Mutant receptors containing nonconservative and conservative substitutions exhibited similar staining patterns as wild-type SR-BI in fixed COS-7 cells  as judged by bright fluorescence at the plasma membrane and cell surface extensions (Figure 3). No staining was observed with non-transfected COS-7 cells (data not shown). Biotinylation experiments also confirmed cell surface expression of all mutant SR-BI proteins (Figure 4A). Immunoblot analysis of SR-BI expression in total cell lysates is also shown (Figure 4B). The lower expression of the V221N mutant receptor in both the total cell lysate and at the cell surface is due to variations in transient expression of the mutant receptor since we have previously observed wild-type or slightly below wild-type levels of total cell lysate expression and cell surface biotinylation in separate experiments (data not shown).
In order to determine the functional consequences of our point mutations in the putative extracellular hydrophobic regions, COS-7 cells were transiently transfected with vectors encoding wild-type or mutant SR-BI. Forty-eight hours post-transfection, cells were assessed for HDL binding and selective uptake of HDL-COE. Analysis of the non-conservative set of mutations revealed four mutant receptors—V67N-, L140Q/L142Q-, V164N- and V221N-SR-BI—with significantly reduced abilities to bind HDL (29%, 40%, 18%, and 12% of wild-type SR-BI binding, respectively; Figure 5A)). A similar pattern was observed for the selective uptake of HDL-COE, with the V67N, L140Q/L142Q, V164N, and V221N mutant receptors being virtually unable to mediate this transport function (only 14%, 4%, 5%, and 6% of wild-type HDL-COE uptake, respectively), despite measurable levels of HDL binding (Figure 5B). Analysis of the conservative set of mutations revealed that the V67L, V164L and V221L mutant receptors regained their ability to mediate wild-type levels of HDL binding and HDL-COE selective uptake once their hydrophobicity was restored. While conservative mutations at positions L140/L142 were unable to improve HDL binding, the L140V/L142V mutant receptor displayed an enhanced ability to mediate selective uptake of HDL-COE (60% greater levels than the L140Q/L142Q receptor, Figure 5B). Receptors with mutations at L359 or L411 displayed wild-type levels of HDL binding and selective uptake of HDL-COE.
In addition to its role in HDL-CE uptake, SR-BI also has the ability to stimulate the release of free cholesterol from cells [41–43]. In order to determine whether our point mutations altered SR-BI-mediated cholesterol efflux, COS-7 cells expressing the mutant receptors were assayed for their ability to promote efflux of cholesterol to HDL. Analysis of the first set of mutations revealed that the V67N, V164N, and V221N mutant receptors displayed significantly reduced levels of efflux to HDL (64%, 28%, and 36%, respectively, of wild-type levels). Restoration of hydrophobicity at V67, V164 and V221 allowed the mutant receptors to regain wild-type levels of efflux (Figure 6). Interestingly, despite an inability to bind HDL, the L140Q/L142Q mutant receptor was able to efflux FC to HDL similar to wild-type SR-BI.
SR-BI can also increase the pool of plasma membrane FC available for oxidation by cholesterol oxidase, as judged by a higher membrane content of cholestenone . To determine whether our point mutations affected the re-distribution of plasma membrane cholesterol, COS-7 cells expressing wild-type and mutant SR-BI were treated with exogenous cholesterol oxidase 48 hours post-transfection. With our non-conservative set of mutations, we observed a pattern that paralleled the HDL binding and HDL-CE selective uptake trends where expression of V67N-, L140Q/L142Q-, V164N-, and V221N-SR-BI significantly decreased the size of the oxidase-sensitive pool of membrane FC (35%, 18%, 54%, and 36% of wild-type levels, respectively). From the conservative set of mutations, the V67L and L140V/L142V mutant receptors displayed a trend for increased cholestenone production, yet the values remained significantly lower than wild-type SR-BI based on our statistical analyses. While the V164L and V221L mutant receptors appeared to regain their ability to increase the oxidase-sensitive pool of FC due to their statistically similar levels of cholestenone production as wild-type SR-BI, they were unable to mediate significant increases in cholestenone in comparison to the V164N and V221N receptors, respectively (Figure 7).
We [44, 45] and others [16, 46–48] have demonstrated the presence of SR-BI oligomers in steroidogenic tissues, as well as in a diverse collection of cell lines. It has been hypothesized that SR-BI oligomerization facilitates cholesterol transport between HDL and the plasma membrane via formation of a “hydrophobic channel” . In order to determine whether the loss of cholesterol transport functions reflected a change in the oligomeric status of SR-BI, we performed native polyacrylamide gel electrophoresis (PAGE) of mutant receptors with no-nconservative substitutions in the presence of PFO. PFO-PAGE is used to evaluate the molecular mass of homo-multimeric membrane protein complexes [37, 49–51]. We found that all mutant receptors existed as dimers and higher order oligomers similar to wild-type SR-BI (data not shown).
Since the identification of SR-BI as the HDL receptor [15, 52], numerous studies have examined the mechanism of the HDL-CE selective uptake process via SR-BI. However, few of these reports have commented on how the topology and/or structural organization of SR-BI at the plasma membrane may influence this crucial step in reverse cholesterol transport. We investigated the role of several putative hydrophobic regions of the extracellular domain of SR-BI by creating amino acid substitutions that modified the overall hydrophobicity of a given region in a predictable manner. In general, non-conservative substitutions in the C-terminal portion of the extracellular domain (L359Q and L411Q) did not have major effects on the various SR-BI-mediated cholesterol functions. However, our functional data revealed that loss of hydrophobicity within the N-terminal half of the extracellular domain (V67N, L140Q/L142Q, V164N or V221N mutations) resulted in an inability of SR-BI to properly: (i) bind HDL and mediate selective uptake of HDL-CE, (ii) promote the release of FC to HDL (with the exception of L140Q/L142Q) and (iii) re-organize plasma membrane cholesterol pools to render them sensitive to exogenous cholesterol oxidase. In contrast, most mutant receptors that retained their hydrophobicity (V67L, V164L and V221L) displayed wild-type levels of SR-BI function in all our assays tested. Taken together, these data suggest that the hydrophobic portion(s) of the N-terminal half of the extracellular domain of SR-BI, namely the region encompassing residues 67 to 221, is critical for facilitating the cholesterol transport functions of SR-BI.
While the majority of our data support a role for N-terminal extracellular hydrophobic regions in regulating SR-BI-mediated cholesterol flux, the L140V/L142V mutant receptor stands alone in that these conservative amino acid substitutions also reduce HDL binding, HDL-CE selective uptake or plasma membrane cholestenone levels. One obvious difference is that the L140V/L142V receptor contains two amino acid substitutions while V67L-, V164L- and V221L-SR-BI contain only single substitutions. This is not a likely source of the discrepancy as individual substitutions (L140V or L142V) do not significantly disrupt the predicted hydrophobicity of the region. Rather, it is probable that one or both of these Leu residues make direct contact with another residue in SR-BI to maintain the receptor in the required conformation. In other words, these residues are not likely to interact with HDL since substituting these residues affects SR-BI activities that are both dependent upon (CE selective uptake) and independent of (oxidase sensitivity , FC efflux ) HDL engagement. A thorough mutagenic analysis of these two amino acid positions is required to fully appreciate the structure/function relationships of L140 and L142.
As described above, we have characterized four novel amino acid positions in the extracellular domain of SR-BI that are critical for HDL binding: V67, L140/L142, V164 and V221. Despite the low levels of HDL binding exhibited by some of these mutant receptors, other functions remained mostly intact. For example, the V164N and V221N mutant receptors maintained accessibility to exogenous cholesterol oxidase, consistent with previous observations that HDL binding is not required for this function of SR-BI . Similarly, the V67N and L140Q/L142Q mutants displayed modest to normal levels of FC efflux to HDL, compared to wild-type SR-BI, thus supporting the notion that SR-BI-mediated FC efflux is mostly independent of direct SR-BI-acceptor interactions . Our data is thus consistent with numerous other studies that describe discrete functional subdomains present in the extracellular domain of SR-BI that mediate the various lipid transport activities [27, 33, 54].
Understanding the conformation and/or topology of the extracellular domain of SR-BI, as well as the changes that might occur upon ligand engagement, is critical to understanding the role this region plays in mediating the various cholesterol transport functions. The mere binding of HDL to SR-BI is not sufficient for lipid transfer to occur. Rather, efficient selective uptake of HDL-CE requires the formation of a “productive complex” [32, 55] where the lipoprotein and receptor must be precisely aligned and/or have the capacity to undergo conformational changes in order to support lipid transfer. How then might these mutations alter “productive complex” formation? It seems likely that the N-terminal half of the extracellular domain of SR-BI physically interacts with the HDL ligand. In addition, we hypothesize that this region may interact with or “dip” into the plasma membrane to facilitate lipid transfer, as has been described for CD36, a member of the class B sub-family of scavenger receptors that shares 70% sequence similarity with SR-BI . The “dipping” of an extracellular domain into a lipid bilayer has also been reported for several other integral membrane proteins including rhomboid protease GlpG  and the Ste2p G-protein-coupled receptor . Therefore, we propose that loss of hydrophobicity in specific extracellular regions of SR-BI alters receptor topology such that proper “productive complex” formation between SR-BI and HDL is prevented.
In conclusion, the N-terminal region of the extracellular domain of SR-BI contains hydrophobic regions critical for receptor-mediated cholesterol flux. We hypothesize that these regions interact with HDL and/or the plasma membrane to facilitate lipid transfer.
The authors thank Rhiannon Ledgerwood for excellent technical assistance. This study was supported by National Institutes of Health grant HL-58012 (to D.S.).
1Abbreviations: ACAT, acyl CoA:cholesterol acyltransferase; CE, cholesteryl ester; COE, cholesteryl oleoyl ether; DLT, dilactitol tyramine; DMEM, Dulbecco’s modified Eagle’s medium; FC, free cholesterol; GAPDH, glyceraldehyde-3-phospate dehydrogenase; HDL, high density lipoprotein; PAGE, polyacrylamide gel electrophoresis; PFO, perfluoro-octanoic acid; SR-BI, scavenger receptor class B type I.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.