HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 47, 36596-36604, November 24, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/47/36596    most recent M006924200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yancey, P. G. Articles by Rothblat, G. H. Search for Related Content PubMed PubMed Citation Articles by Yancey, P. G. Articles by Rothblat, G. H. Social Bookmarking                      What's this?

High Density Lipoprotein Phospholipid Composition Is a Major Determinant of the Bi-directional Flux and Net Movement of Cellular Free Cholesterol Mediated by Scavenger Receptor BI^*

Patricia G. Yancey, Margarita de la Llera-Moya, Snehasikta Swarnakar^§, Pascale Monzo^§, Seth M. Klein^§, Margery A. Connelly^§, William J. Johnson, David L. Williams^§¶, and George H. Rothblat

From the  Division of Gastroenterology and Nutrition, Department of Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania 19104 and the ^§ Department of Pharmacological Sciences, University Medical Center, State University of New York, Stony Brook, New York 11794

Received for publication, August 1, 2000, and in revised form, August 11, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The role of high density lipoprotein (HDL) phospholipid in scavenger receptor BI (SR-BI)-mediated free cholesterol flux was examined by manipulating HDL₃ phosphatidylcholine and sphingomyelin content. Both phosphatidylcholine and sphingomyelin enrichment of HDL enhanced the net efflux of cholesterol from SR-BI-expressing COS-7 cells but by two different mechanisms. Phosphatidylcholine enrichment of HDL increased efflux, whereas sphingomyelin enrichment decreased influx of HDL cholesterol. Although similar trends were observed in control (vector-transfected) COS-7 cells, SR-BI overexpression amplified the effects of phosphatidylcholine and sphingomyelin enrichment of HDL 25- and 2.8-fold, respectively. By using both phosphatidylcholine-enriched and phospholipase A₂-treated HDL to obtain HDL with a graded phosphatidylcholine content, we showed that SR-BI-mediated cholesterol efflux was highly correlated (r² = 0.985) with HDL phosphatidylcholine content. The effects of varying HDL phospholipid composition on SR-BI-mediated free cholesterol flux were not correlated with changes in either the K_d or B_max values for high affinity binding to SR-BI. We conclude that SR-BI-mediated free cholesterol flux is highly sensitive to HDL phospholipid composition. Thus, factors that regulate cellular SR-BI expression and the local modification of HDL phospholipid composition will have a large impact on reverse cholesterol transport.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The deposition of cholesterol in peripheral cells is opposed by the process of reverse cholesterol transport (RCT)¹ where high density lipoproteins (HDL) remove free cholesterol (FC) from cells and deliver it back to the liver for excretion (1-3). The flux of FC between cells and HDL is bi-directional. Depending on the direction of the FC concentration gradient between cells and lipoproteins, either net efflux or net influx of cholesterol can occur (4, 5). The creation of a cholesterol gradient depends upon many properties of the acceptors and the cell plasma membrane. Such factors include the cholesterol and phospholipid content of the acceptors and plasma membrane (4, 5), the existence of cholesterol domains within the plasma membrane (6-10), and the size, number, and composition of acceptor particles (11-13).

Recent studies have shown that, when cells express scavenger receptor BI (SR-BI), the bi-directional flux of FC between cells and HDL is accelerated (8, 14, 15). The mechanism by which SR-BI mediates FC flux is uncertain. However, recent studies from our laboratory demonstrated that binding of the acceptor particles to SR-BI is not a requirement for SR-BI-mediated cholesterol efflux (7, 8). Rather SR-BI induces a reorganization of the plasma membrane cholesterol, and this reorganization is linked to enhanced FC flux (7, 8, 16). Regardless of the mechanism, evidence is accumulating to support the importance of SR-BI-mediated FC flux in RCT. Recent studies of Ji and colleagues (17) showed that either attenuation or overexpression of hepatic SR-BI in mice led to significantly decreased or increased delivery of HDL FC into bile. In addition, the expression of SR-BI in peripheral cells and in foam cells of the arterial wall suggests a role for SR-BI in the removal of FC from the periphery (15, 18, 19).

SR-BI-mediated FC flux requires phospholipid in the acceptor (15), and studies have shown that cholesterol efflux from cells is highly correlated with the concentration of HDL phospholipid in serum (20, 21). Also, the stimulation of cholesterol efflux upon phospholipid supplementation of serum is closely linked to the levels of SR-BI among cell types (14). These observations are consistent with epidemiological data demonstrating that humans with low HDL phospholipid levels have a high incidence of coronary artery disease (22). These findings suggest that changes in HDL phospholipid content may alter SR-BI-mediated FC flux. The current studies explicitly test this hypothesis by determining the effects of manipulating HDL phosphatidylcholine and sphingomyelin content on both the influx and efflux of FC using SR-BI-expressing COS-7 cells. Our results demonstrate that SR-BI-mediated bi-directional FC flux is highly sensitive to HDL phospholipid content and composition.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Tissue culture plasticware was obtained through Falcon (Lincoln, NJ). Calf serum (CS), bovine serum albumin (BSA), unesterified cholesterol, cholesteryl methyl ether, penicillin, streptomycin, and phospholipase A₂ (PLA₂, from Crotalus adamanteus venom, P-0790) were purchased from Sigma. [1,2-³H]Cholesterol was purchased from PerkinElmer Life Sciences. Bovine brain sphingomyelin, 1,2-dimyristoyl-sn-glycerophosphocholine (DMPC), 1-palmitoyl-2-oleoyl-sn-glycerophosphoserine, (POPS), and 1-palmitoyl-2-oleoyl-sn-glycerophosphocholine (POPC) were obtained from Avanti Polar Lipids (Alabaster, AL). All other reagents and organic solvents were purchased from Fisher. The acyl-CoA:cholesterol acyltransferase inhibitor, compound CP113,818, was a generous gift from Pfizer.

Phospholipid Vesicle Preparation and HDL Phospholipid Modification-- POPC small unilamellar vesicles (SUV) containing 15 mol % FC were made as described previously (23). Similar procedures were used to make SUV containing 15 mol % cholesterol and also both POPC and POPS (50% mol/mol). HDL₂ (1.066 g/ml  d  1.125 g/ml) and HDL₃ (1.125 g/ml  d  1.210 g/ml) were isolated by sequential ultracentrifugation (24). Contaminating LDL was removed from HDL₂ by heparin-Sepharose column chromatography as described previously (25). Prior to use, HDL was dialyzed extensively against 0.9% NaCl, 10 mM HEPES (pH 7.4) and sterilized by filtration through a 0.45-µm Millipore filter.

To enrich HDL₃ with phospholipid, DMPC, or sphingomyelin, multilamellar vesicles (MLVs) were prepared as described (26). For enrichment with DMPC, 1-2 mg of HDL₃ protein/ml was incubated for 2 h at 24 °C with DMPC MLVs added at doses that ranged from 0.25- to 3-fold the total native amount of HDL₃ phospholipid. For enrichment with sphingomyelin, MLVs were added in amounts ranging from 0.5- to 20-fold the HDL₃ sphingomyelin content (estimated to be 12% of HDL₃ total phospholipid), and the mixture was incubated for 1 h at 37 °C. Since the phase transition temperature of sphingomyelin is broad, the mixture was then warmed to 42 °C and allowed to slowly cool to 25 °C over a 5-h period. During enrichment with either DMPC or sphingomyelin, control HDL₃ was incubated similarly but without adding MLVs. After incubation of the HDL with MLVs, any unreacted MLVs were removed by sequentially filtering the HDL MLV mixture through 0.45- and 0.22-µm filters. After overnight storage at 4 °C, any remaining MLVs were removed from HDL by centrifugation for 30 min at 3,000 × g. The phospholipid to protein ratio of control HDL₃ used in these studies ranged from 0.27 to 0.48, and the degree of enrichment for the different HDL preparations is indicated under "Results."

Another approach to manipulating the PC content of HDL₃ was PLA₂ treatment. HDL₃ or DMPC-enriched HDL₃ (1-2 mg of protein/ml) in buffer containing 20 mM Tris, 0.15 M NaCl, 8 mM CaCl₂, and 1% BSA (w/v) was incubated at 37 °C from 1.5 to 20 min with 1 µg of PLA₂/ml. The reaction was stopped by the addition of 16 mM EDTA. To remove the enzyme and lyso-PC albumin complexes, the HDL₃ was isolated by ultracentrifugation for 24 h at d = 1.210 g/ml. Control HDL₃ was treated similarly as PLA₂-treated HDL₃ except without the addition of the enzyme. To remove EDTA, the HDL was dialyzed extensively against 0.15 M NaCl containing 10 mM HEPES.

Cell Culture and Transient Transfections-- COS-7 cells were maintained on DMEM supplemented with 10% CS and antibiotics. For transfection, cells were seeded in 100-mm plates and incubated for 18 h at 37 °C in DMEM supplemented with 10% CS. Cells were transfected with 10 µg of the indicated plasmid and diluted in serum-free DMEM and Fugene 6 (Roche Molecular Biochemicals) as described previously (8). The plasmid containing mixture was added dropwise to the plated cells. The pSG5 vector (Stratagene, Inc.) with or without murine SR-BI or rat CD36 were prepared using endotoxin-free Qiagen Maxiprep kits (8, 27).

Measurement of Cholesterol Efflux and Influx-- After transfection, the cells were removed from the 100-mm plates by trypsinization. The transfected cells were suspended in 10% CS/DMEM containing 2 µg of CP113,818/ml and replated into 12-well plates. One 100-mm plate yielded one 12-well plate. For cholesterol efflux, 1 ml of 10% CS/DMEM containing 6 µCi of [³H]cholesterol/ml and 2 µg of CP113,818/ml was added to the transfected cells immediately after replating. After 24 h of incubation with the labeling medium, the cells were washed once with 1% BSA/DMEM and once with DMEM. Medium containing HDL or phospholipid vesicles at the desired concentration was then added to the cells and incubated at 37 °C for varying times. At the indicated time points, 150-µl aliquots of the medium were taken and filtered through 0.45-µm Multiscreen filtration plates. The [³H]cholesterol in 100 µl of filtrate was measured by liquid scintillation counting. In all experiments fractional efflux was corrected for the small amount of [³H]cholesterol released to DMEM without HDL present. Cholesterol influx was measured by using transfected cells prepared in exactly the same way as for efflux, except [³H]cholesterol was not included in the medium. To measure influx, HDL or phospholipid vesicles were labeled by exchanging [³H]cholesterol (20-40 µCi/mg HDL protein or 25 µCi/mg vesicle phospholipid) from the glass wall of a test tube onto which the [³H]cholesterol had been dried under N₂. After incubation of the HDL or vesicles with the [³H]cholesterol overnight at 4 °C, the particles were sterilized by filtration through a 0.45 µm filter. The radiolabeled HDL or vesicles were diluted in DMEM and incubated with the unlabeled cells. At each time point, the cells were washed three times with PBS and the cell lipids extracted with isopropyl alcohol. The total [³H]cholesterol present in the total lipid extract was measured by liquid scintillation counting.

Analytical Procedures-- The unesterified and esterified cholesterol contents of HDL and cells were measured by gas-liquid chromatography as described previously (28). Cell and HDL protein determinations were done using the method of Lowry et al. (29) as modified by Markwell et al. (30). HDL phospholipid was determined by the method of Sokoloff and Rothblat (31). The sizes of native and modified HDL samples were determined by gel exclusion chromatography on a 25-ml Superose 6 column (Amersham Pharmacia Biotech). Briefly, 300 µg of HDL protein in a volume of 400 µl were run at 0.4 ml/min in 10 mM potassium phosphate buffer (pH 7.2), 150 mM NaCl. HDL elution was monitored by absorbance at 280 nm.

Kinetic Analysis-- The transfer of FC between cells and HDL has been shown to be a bi-directional process. In the current studies, the kinetic analysis assumes a closed system in which FC exists in one of two pools, either the HDL pool or the cellular pool, and the analysis of flux data was accomplished as described previously (4). In our experiments, the kinetic analysis of the bi-directional flux of FC is not affected by the esterification of cholesterol, as acyl-CoA:cholesterol acyltransferase was inhibited during all incubations. A single exponential equation that describes the bi-directional transfer between two pools was fitted to the data by computer (Prism, GraphPad Inc., San Diego, CA). The equation is as follows: Y = H₁e^gt + H₂. Y represents either the fractional uptake of HDL [³H]cholesterol or the fractional retention of [³H]cholesterol by the cells, and t is the incubation time in hours. H₁, g, and H₂ are constants adjusted by the program to fit the equation to the data. The constant g is the sum of the rate constants of efflux (k_e) and influx (k_i) in fraction/h. The values k_e and k_i are the initial slopes (t = 0) of the curves describing [³H]cholesterol retention or uptake, respectively. The initial unidirectional efflux (F_e) or influx (F_i) of FC mass in µg of FC/(h × mg cell protein) are estimated as follows: F_e = k_e × cell FC mass/mg cell protein and F_i = k_i × HDL FC mass in total volume of medium/mg cell protein. The initial net flux of FC mass equals F_i  F_e. It should be noted that the estimates of mass flux do not take into account the contribution of SR-BI-mediated selective uptake of HDL cholesteryl ester (16, 27, 32).

Radioiodination of Native and Modified HDL and Analysis of HDL Binding to SR-BI-- Native and modified HDL₃ particles were dialyzed against PBS, 0.25 mM EDTA (pH 7.4), and iodinated with Na¹²⁵I using the modified ICL method as described (33). The sample was then passed through a PD-10 column equilibrated with PBS, 0.25 mM EDTA (pH 7.4), and dialyzed against PBS, 0.25 mM EDTA, 100 mM KI (pH 7.4), followed by 3 changes of PBS, 0.25 mM EDTA (pH 7.4) at 4 °C. The recovery of iodinated particles was 80-90%. The specific activities for various HDL particles ranged from 146 to 670 cpm/ng of protein.

For binding analysis, COS-7 cells were transfected with SR-BI expression plasmid or vector as described above and replated in 24-well plates. Twenty four hours later, the cells were washed 3 times with HEPES-buffered Eagle's minimal essential medium containing 1% BSA. Native or modified HDL particles were then added in the same medium (0.5 ml) to triplicate wells, and the cells were incubated for 90 min at 37 °C. At the end of the incubation, the medium was removed and centrifuged at 10,000 × g for 5 min, and an aliquot was used for -counting to determine the free ligand concentration. Cells were placed on ice and washed three times with cold medium containing 1% BSA and three times with cold PBS (pH 7.4), followed by solubilization in 0.5 ml of 0.1 NaOH at room temperature. The sample was transferred to a vial for -counting after which it was used for protein determination (29). SR-BI-specific binding of native or modified HDL was determined by subtracting values for vector-transfected cells from SR-BI-expressing cells to generate a SR-BI-vector curve. Binding parameters for B_max and K_d values were obtained for the SR-BI vector curve by nonlinear regression (Prism, GraphPad Inc., San Diego, CA) using a one-site binding isotherm as indicated (34, 35).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Comparison of HDL₂ and HDL₃ as FC Acceptors-- Since HDL₂ is enriched in phospholipid in comparison to HDL₃ (36), initial studies compared cholesterol efflux from control or SR-BI-expressing COS-7 cells to the two lipoprotein particles (Fig. 1, A and B). At all HDL₂ or HDL₃ protein concentrations, FC efflux was 5- to 7-fold greater in SR-BI-expressing cells compared with control cells. When the data were normalized to HDL protein, HDL₂ was more efficient than HDL₃ at stimulating efflux in both SR-BI-expressing (Fig. 1A) and control cells (Fig. 1B). However, this difference was more pronounced in SR-BI-expressing cells than in control cells. When the efflux values were normalized to HDL phospholipid content (Fig. 1, C and D), there was no difference in efflux efficiency between HDL₂ and HDL₃ in either control or SR-BI-expressing cells. This demonstrates that the difference in efficiency seen with similar protein concentrations is due to HDL₂ being more enriched with phospholipid compared with HDL₃.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Cholesterol efflux from control and SR-BI-expressing cells to HDL2 and HDL3. SR-BI-expressing (A and C) and control (B and D) cells were labeled with [3H]cholesterol as described under "Experimental Procedures." After cholesterol labeling, the cells were incubated for 2 h at 37 °C with the indicated HDL2 () or HDL3 () protein concentration. Shown is the % cholesterol efflux/2 h plotted against HDL protein (A and B) or HDL phospholipid (C and D) content. The values are means ± S.D. of triplicate determinations. The phospholipid to protein ratios were 0.83 ± 0.01 and 0.36 ± 0.01 (w/w) for HDL2 and HDL3, respectively.

Enrichment of HDL₃ with PC-- We tested the role of HDL phospholipid in SR-BI-mediated FC flux by enriching HDL₃ with DMPC. Shown in Fig. 2, A and B, is the time courses of efflux and influx with SR-BI-expressing cells incubated with native HDL₃ or DMPC-enriched HDL₃. There was substantially more cholesterol efflux with DMPC-enriched HDL compared with native HDL (Fig. 2A). In contrast, there was no effect of DMPC enrichment on the initial rate of cholesterol influx with SR-BI-expressing cells (Fig. 2B). For subsequent experiments, rate constants for efflux (k_e) and influx (k_i) were calculated by curve fitting (see "Experimental Procedures") and are presented as k_e and k_i in fraction/h.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Time courses of the bi-directional flux of FC between SR-BI-expressing cells and either native HDL3 or PC-enriched HDL3. A, fractional efflux of cholesterol. After [3H]cholesterol labeling, cells were incubated for up to 6 h at 37 °C with DMEM alone or DMEM containing native HDL3 () or PC-enriched HDL3 () (both at 20 µg of protein/ml). The values are the means ± S.D. of triplicate determinations. B, fractional uptake of cholesterol. Cells were incubated for up to 6 h at 37 °C with [3H]cholesterol-labeled native HDL3 or PC-enriched HDL3 (both at 20 µg protein/ml). The values are the mean ± S.D. of triplicate determinations. The native and PC-enriched HDL phospholipid to protein ratios were 0.26 ± 0.02 and 1.1 ± 0.02 (w/w), respectively.

Incubation of control COS-7 cells with HDL₃ (20 µg of protein/ml) that had been enriched with increasing levels of DMPC showed a small effect of HDL₃ PC enrichment (slope = 0.0025 ± 0.0001) on efflux (Fig. 3A). However, there was a 25-fold greater effect of HDL₃ PC enrichment on SR-BI-mediated efflux (slope = 0.0604 ± 0.003). Unlike SR-BI-mediated efflux, there was no effect of HDL₃ PC enrichment on SR-BI-mediated influx, and similarly, no effect was seen with control cells (Fig. 3B). Similar results were observed when SR-BI-expressing or control cells were incubated with either HDL₃ or PC-enriched HDL₃ at 200 µg of HDL protein/ml (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Effects of enriching HDL3 with increasing levels of PC on the bi-directional flux of FC with SR-BI-expressing or control cells. A, efflux of cholesterol. After [3H]cholesterol labeling, cells were incubated for up to 6 h at 37 °C in DMEM alone or with 20 µg of protein/ml of native HDL3 or HDL3 preparations enriched with increasing levels of PC. The efflux data are presented as ke in units of fraction/h and are derived from time courses with triplicate determinations at each time point. , SR-BI-expressing cells; , control cells. B, for influx, parallel sets of unlabeled cells were incubated for up to 6 h at 37 °C with 20 µg of protein/ml of [3H]cholesterol-labeled native HDL3 or HDL3 preparations enriched with increasing levels of PC. The data are presented as ki in units of fraction/h and are calculated from influx time course data with triplicate determinations at each time point.

Depletion of HDL₃ PC with PLA₂-- We next examined the role of HDL PC in SR-BI-mediated cholesterol flux by decreasing the PC content of HDL₃ with phospholipase A₂ treatment. As shown in Fig. 4A, depletion of HDL₃ PC had a large effect on SR-BI-mediated cholesterol efflux, decreasing the rate constant (k_e) by 70%. Depletion of HDL PC also decreased the rate of cholesterol influx in SR-BI-expressing cells (Fig. 4B). Unlike SR-BI-expressing cells, depletion of HDL₃ PC resulted in little or no change in the rates of FC efflux and influx in control cells. Similar results were observed when SR-BI-expressing or control cells were incubated with HDL₃ and PC-depleted HDL₃ at 200 µg of HDL protein/ml (data not shown). The use of a combination of DMPC enrichment and PLA₂ treatment to obtain HDL₃ with a graded PC content showed that, regardless of the means used to manipulate HDL₃ PC content, a strong relationship (r² = 0.985) existed between SR-BI-mediated efflux and HDL₃ PC content (Fig. 5A). In contrast, there was no relationship (r² = 0.0002) between efflux and HDL PC content in control cells (Fig. 5B). There was also no correlation between SR-BI-mediated FC influx (r² = 0.172) and HDL₃ PC content (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effects of depleting HDL PC content on the bi-directional flux of FC with SR-BI-expressing and control cells. A, efflux of cholesterol. After [3H]cholesterol labeling, the cells were incubated for up to 6 h at 37 °C in DMEM alone or with 20 µg of protein/ml of native HDL3 or HDL3 depleted of PC. The efflux data are presented as ke in units of fraction/h and are the means ± S.D. of triplicate determinations. B, for influx, parallel sets of unlabeled cells were incubated for up to 6 h at 37 °C with 20 µg of protein/ml [3H]cholesterol-labeled native HDL3 or HDL3 depleted of PC. The influx data are presented as ki in units of fraction/h and are the means ± S.D. of triplicate determinations. The HDL phospholipid to protein ratios were 0.27 ± 0.02 and 0.08 ± 0.02 (w/w) for native HDL3 or HDL3 depleted of PC.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effects of HDL3 with graded PC content on the efflux of FC from SR-BI-expressing (A) and control (B) cells. After [3H]cholesterol labeling, the cells were incubated for up to 6 h at 37 °C in DMEM alone or with 20 µg of protein/ml of the following: native HDL3 (), native HDL3 treated for 1.5 min with PLA2 (), native HDL3 treated for 20 min with PLA2 (), PC-enriched HDL3 (), PC-enriched HDL3 treated for 1.5 min with PLA2 (), or PC-enriched HDL3 treated for 20 min with PLA2 (). The efflux data are presented as ke in units of fraction/h and are derived from time courses with triplicate determinations at each time point.

Enrichment of HDL₃ with Sphingomyelin-- We next examined the effect of HDL sphingomyelin on SR-BI-mediated flux. Shown in Fig. 6, A and B, are k_e and k_i when SR-BI-expressing or control cells were incubated with HDL₃ that had been enriched with increasing levels of sphingomyelin. In contrast to PC enrichment of HDL₃ (see Figs. 2, 3, and 5), sphingomyelin enrichment had only a small effect on SR-BI-mediated efflux (Fig. 6A) but a large effect on cholesterol influx (Fig. 6B). Similar to SR-BI cells, there was only a minimal effect of HDL₃ sphingomyelin enrichment on efflux in control cells but the rate of influx decreased. To compare the effects of sphingomyelin enrichment of HDL in control and SR-BI-expressing cells, the dependence of k_i on sphingomyelin content was linearized by plotting the log of k_i against the HDL phospholipid to protein ratio (Fig. 6C). When plotted this way, the difference in slopes (0.88 versus 0.44) indicates that influx to SR-BI cells is 2.8 (=10^(0.88-0.44)) times more sensitive to HDL sphingomyelin enrichment than influx to control cells. Similar results were observed when SR-BI-expressing or control cells were incubated with control and sphingomyelin-enriched HDL at 200 µg of HDL protein/ml (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Effects of enrichment of HDL3 with sphingomyelin on the bi-directional flux of FC with SR-BI-expressing or control cells. A, efflux of cholesterol. After [3H]cholesterol labeling, the cells were incubated for up to 6 h at 37 °C in DMEM alone or with 20 µg of protein/ml of native HDL3 or HDL3 enriched with increasing levels of sphingomyelin. The efflux data are presented as ke in units of fraction/h and are derived from time courses with triplicate determinations at each time point. , SR-BI-expressing cells; , control cells. B, for influx, parallel sets of unlabeled cells were incubated for up to 6 h at 37 °C with 20 µg of protein/ml of [3H]cholesterol-labeled native HDL or HDL enriched with increasing levels of sphingomyelin. The data are presented as ki in units of fraction/h and are calculated from influx time course data with triplicate determinations at each time point. C, shown are the log of the ki values from B plotted against the HDL phospholipid to protein ratio. The lines are linear regression fits to the data.

Effect of Phospholipid Enrichment or Depletion on HDL Size-- To determine whether enrichment of HDL with PC or sphingomyelin or depletion of PC by phospholipase treatment significantly altered the size of the HDL, native and modified HDL were analyzed by gel exclusion chromatography. None of these treatments significantly altered the chromatographic profile of HDL (data not shown). Scanu and colleagues (37) also reported that severe depletion of HDL PC by phospholipase treatment did not alter HDL size.

Calculations of Net Flux-- The initial rate constants for efflux and influx from Figs. 3 and 6 were used to estimate the initial net flux of FC mass from SR-BI-expressing and control cells incubated with PC or sphingomyelin-enriched HDL₃ (Fig. 7, A and B). Enrichment of HDL₃ with either PC or sphingomyelin led to enhancements in the net efflux of FC mass from SR-BI-expressing cells. In contrast, there were no changes in net flux of FC mass from control cells incubated with PC or sphingomyelin-enriched HDL. The ratio of k_e to k_i predicts the steady-state distribution of FC between the HDL and cells after long incubation times (Fig. 8, A and B) (4). These predictions follow the same trends as the estimates of net flux. In SR-BI-expressing cells incubated with native HDL, the FC was minimally redistributed toward the HDL compared with control cells. The shift in FC to HDL in SR-BI-expressing cells was greatly enhanced upon enrichment of HDL₃ with increasing levels of either PC or sphingomyelin. These effects were much less pronounced with control cells.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Effects of DMPC or sphingomyelin enrichment of HDL3 on the net flux of FC with SR-BI-expressing () and control () cells. The rate constants from Figs. 3 and 6 were used to calculate the initial net flux of FC as described under "Experimental Procedures." The data are expressed as µg of FC/(h × mg cell protein).

View larger version (20K): [in this window] [in a new window]   Fig. 8.   The effect of DMPC (A) or sphingomyelin (B) enrichment of HDL3 on the estimated steady-state distribution of FC. The ke values from Figs. 3 and 6 were divided by the ki values from Figs. 3 and 6 to estimate the ratio of medium FC to cellular FC at steady state. , SR-BI-expressing cells; , control cells.

Effect of HDL Phospholipid Composition on Binding to SR-BI-- We next tested whether differences in binding of the various modified HDLs to SR-BI-expressing cells could account for the changes in SR-BI-mediated FC flux. Shown in Fig. 9, A and B, are representative binding curves for ¹²⁵I-labeled native HDL₃ and PC-enriched/PLA₂-depleted HDL₃ (see Fig. 9 legend) to control and SR-BI-expressing cells. SR-BI-specific binding (dashed curves) was similar between native and the modified HDL particles. The K_d values (µg of HDL protein/ml) were 8.0 ± 1.4 and 9.0 ± 1.9, and the B_max values (ng of HDL protein/mg cell protein) were 127 ± 7 and 127 ± 9 for native and modified HDL, respectively. Summarized in Table I are the mean K_d values for replicate experiments with the different HDL preparations. The only significant change in affinity for binding to SR-BI was with the phospholipase A₂-treated native HDLs where the affinity decreased by 56-61% compared with native HDL₃. However, it is clear that the dramatic increase in FC efflux with PC-enriched HDL and the decrease in FC influx with sphingomyelin-enriched HDL are not due to altered binding to SR-BI since there are no differences in K_d values among these particles.

View larger version (25K): [in this window] [in a new window]   Fig. 9.   Binding of native HDL3 and DMPC-enriched/PLA2-treated HDL to SR-BI. SR-BI-expressing () and vector-transfected () COS-7 cells were incubated with 125I-native HDL3 (A) or with modified 125I-HDL3 (enriched with DMPC and then treated with PLA2 for 20 min) (B) for 90 min at 37 °C as described under "Experimental Procedures." Bound HDL3 (ng of HDL protein/mg cell protein) is plotted versus the concentration of free HDL measured at the end of the incubation. The SR-BI vector curve in each panel (dashed line) was obtained by subtracting vector values for each ligand concentration from the corresponding SR-BI values and fitting the resultant data via nonlinear regression using a one-site binding isotherm. Error bars show the S.D. of triplicate values. Phospholipid/protein (w/w) ratios were 0.38 ± 0.02 for native HDL3 and 0.60 ± 0.02 for modified HDL3.

Previous studies (7, 8) from our laboratory suggested that binding of acceptor particles to SR-BI was not required for SR-BI-mediated FC efflux. We next determined if this is also the case for SR-BI-mediated FC influx by comparing the bi-directional flux of FC using neutral (POPC) or charged (POPC:POPS) vesicles with SR-BI-expressing and control cells (Table II). Vesicles containing 50 mol % POPS have been shown to bind to SR-BI, whereas vesicles composed of only POPC do not bind to SR-BI (38, 39). Influx of FC from both POPC and POPC:POPS vesicles was stimulated similarly when SR-BI-expressing cells were compared with control cells. This suggests that binding to SR-BI is not a requirement for SR-BI-mediated influx. In addition, similar to our previous results (8), SR-BI-mediated efflux was stimulated more with POPC vesicles as acceptors compared with POPC:POPS vesicles. Thus, these data indicate that binding of acceptor/donor particles to SR-BI cannot explain the effect of SR-BI on bi-directional FC flux.

View this table: [in this window] [in a new window]   Table II Bi-directional flux of free cholesterol between neutral or charged phospholipid vesicles and COS cells +/ SR-BI Rate constants for influx and efflux of FC with control or SR-BI expressing cells incubated with 1 mg of phospholipid/ml of either small unilamellar vesicles (SUV) containing 15 mol % cholesterol and 85 mol % phosphatidylcholine (PC) or vesicles containing 15 mol % cholesterol, 42.5 mol % PC, and 42.5 mol % phosphatidylserine (PS) were determined as described under "Experimental Procedures." Values are means ± S.D.

We also examined whether influx was stimulated in cells expressing CD36, another scavenger receptor to which HDL binds with high affinity (27, 40). Consistent with the conclusion that HDL binding, per se, is not sufficient to enhance influx of HDL FC, the rate constants (fraction/h) for FC influx from HDL were 0.064 ± 0.006, 0.067 ± 0.005, and 0.170 ± for control, CD36-expressing, and SR-BI-expressing cells, respectively. Thus, similar to previous studies on FC efflux (8), the simple concentration of HDL particles at the cell membrane via tethering to CD36 does not explain the greater influx of FC seen with SR-BI-expressing cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Recent studies have shown that the bi-directional flux of FC is accelerated in cells expressing SR-BI (8, 15). The importance of SR-BI-mediated FC flux in RCT is substantiated by studies showing that manipulations in hepatic SR-BI expression result in parallel changes in both removal of HDL FC from plasma and incorporation of HDL FC into bile (17). In addition, other studies have shown that SR-BI is present in foam cells of atherosclerotic lesions (17, 19). The current studies demonstrate that SR-BI-mediated FC flux is very sensitive to the PC and sphingomyelin composition of HDL and that HDL phospholipid composition alters the steady-state distribution of cholesterol between SR-BI-expressing cells and HDL (Fig. 7).

The Effect of HDL PC and Sphingomyelin Content on SR-BI-mediated FC Flux-- Enrichment of HDL₃ with DMPC markedly enhanced SR-BI-mediated cholesterol efflux (Fig. 3). This stimulation in SR-BI-mediated efflux was not specific to DMPC enrichment since decreasing native HDL₃ PC content by treatment with PLA₂ substantially decreased the rate of SR-BI-mediated efflux (Fig. 4A). Furthermore, the difference in SR-BI-mediated efflux to HDL₂ and HDL₃ was explained by the phospholipid content of these native HDL particles (Fig. 1, C and D). In addition, SR-BI-mediated efflux was correlated with HDL PC content over a wide range of HDL phospholipid to protein ratios (Fig. 5). These findings explain prior studies in which cholesterol efflux was closely correlated to HDL phospholipid content, and where the extent of this phospholipid effect was correlated with SR-BI levels (20, 21, 26).

Unlike the effect of PC enrichment of HDL on SR-BI-mediated cholesterol efflux (Fig. 3), enrichment with sphingomyelin had only a minor effect on efflux but caused a large decrease in SR-BI-mediated cholesterol influx (Fig. 6). The finding that sphingomyelin enrichment of HDL resulted in only a minor stimulation of efflux was surprising since our previous studies showed that sphingomyelin enrichment of whole serum stimulates efflux from Fu5AH cells, which are rich in SR-BI (14). One possible explanation is that in the present investigation, the initial rates of bi-directional flux were determined, whereas our earlier studies measured efflux at a later time that would also reflect the back flux of cholesterol from medium to cells. This back flux would be influenced by the effect of the sphingomyelin enrichment of HDL, which reduces influx. Another possible explanation for this difference is that enrichment of serum promotes the formation of more efficient cholesterol acceptors, as opposed to just the enrichment of HDL₃ alone, as was done in the present investigation.

Both PC and sphingomyelin enrichment of HDL led to substantial enhancements in the estimated net efflux of FC from SR-BI-expressing cells but had only a small effect on net flux from control cells (Fig. 6). Thus, both types of HDL phospholipid manipulations led to similar effects on cell FC mass, but by different mechanisms. Sphingomyelin enrichment of HDL enhanced net efflux by decreasing SR-BI-mediated influx, whereas PC enrichment of HDL increased SR-BI-mediated efflux.

Mechanisms for SR-BI-mediated FC Flux-- The mechanism by which SR-BI stimulates FC flux remains unresolved. It is conceivable that manipulating the lipid composition of HDL could result in changes in the binding of the lipoprotein to SR-BI. Previous studies from this laboratory have shown that SR-BI expression stimulates efflux of cell cholesterol to acceptors that have been reported not to bind to SR-BI (7, 8). The present studies have extended these observations and also suggest that neither efflux nor influx of cholesterol is a function of specific, high affinity binding of the type that has been shown to promote selective uptake of cholesteryl esters from HDL (27, 34, 41). Consistent with this conclusion are the following observations. 1) Cholesterol influx studies using both neutral SUV donors, which do not bind to SRBI, and negatively charged vesicles, which do bind, demonstrated similar enhancements of influx upon expression of the receptor (Table II). 2) Enrichment of HDL₃ with either PC or sphingomyelin changed FC flux without changing the K_d or B_max values for binding of the modified particles to SR-BI (Table I and Figs. 3 and 6). 3) PC and sphingomyelin enrichment of HDL did not produce similar quantitative changes in both influx and efflux of FC (Figs. 3, 6, and 9).

In contrast to the observations discussed above, one set of modified HDL particles did show a parallel relationship between flux and binding. Thus, HDL severely depleted in PC (i.e. 79% reduction) exhibited reduced influx and efflux (Fig. 4), together with reduced binding affinity (Table I). However, the change in binding affinity was minor (2-fold), and the same relative changes in efflux were seen at 200 µg/ml HDL protein at which concentration SR-BI would be essentially saturated with bound HDL whether the K_d was 6 or 15 µg/ml. Thus, the change in binding affinity is unlikely to explain the changes in flux seen with PLA₂-treated HDL. More likely the changes in FC flux may be related to the extensive removal of PC which would produce an HDL relatively enriched in sphingomyelin that would be expected to exhibit both reduced efflux and influx of FC. In addition, similar to the present results, studies of Lagrost and colleagues (42) showed that FC efflux from SR-BI-rich hepatoma cells is decreased when the apolipoprotein AI (apoAI) content of HDL is replaced by apoAII, even though other studies have shown that the binding affinity of these apoAII-enriched particles is higher compared with native HDL (43). Furthermore, a recent study by Chen et al. (44) demonstrated that an antibody directed against SR-BI which blocks both HDL binding and selective cholesteryl ester uptake does not reduce the efflux of cellular FC to HDL.

In contrast to the studies discussed above, recent studies of Gu and colleagues (45) are consistent with binding of HDL to SR-BI being essential for SR-BI-mediated FC efflux. It was reported that antibody to SR-BI does block FC efflux to HDL and that mutations that block HDL binding to SR-BI also block FC efflux (45). The basis for these differences is unclear at present, although a number of possibilities may be suggested. For example, PC enrichment of HDL in the present study might create a small subset of particles that bind to SR-BI with higher affinity and are responsible for the enhanced SR-BI-mediated FC efflux. It seems unlikely to us that a small subset of particles, undetectable by binding analysis, could explain the 25-fold increase in SR-BI-mediated FC efflux, but this possibility cannot be excluded. Another possibility, previously suggested (8, 16), is that SR-BI-mediated FC efflux has two components, one of which requires HDL binding to SR-BI, whereas the other, which occurs with neutral phospholipid vesicles (8) and cyclodextrin acceptors (7), may reflect changes in plasma membrane lipid domains and be independent of acceptor binding. Although these points are currently unresolved, it is clear from the studies with CD36 that simple binding of HDL to the cell surface is not sufficient to enhance the bi-directional flux of FC between HDL and cells.

The primary mechanism of FC flux with control cells would be aqueous diffusion (46) since COS-7 cells have minimal background levels of SR-BI (27) and ATP binding cassette transporter-1 (ABC-1) (58). Recent studies from our laboratory suggest that SR-BI enhances cholesterol efflux by reorganizing cholesterol domains in the plasma membrane resulting in areas of the membrane having accelerated rates of cholesterol desorption (7, 8). This conclusion comes from comparative studies on SR-BI-positive and control cells showing that SR-BI expression results in increased sensitivity of plasma membrane cholesterol to cholesterol oxidase and in an increased size of the fast pool of FC available for cyclodextrin-mediated efflux. Presumably, SR-BI reorganization of the membrane also creates domains where FC molecules released from HDL can be readily incorporated into the membrane. Whether these domains of the membrane are the same as the "hydrophobic pathway" (34) or caveolae that are proposed to be involved in SR-BI-mediated selective uptake of cholesteryl ester is unclear at this time (47, 48).

The stimulation in efflux observed with PC enrichment of HDL is likely due to the excess phospholipid solubilizing the desorbed cholesterol molecules, and the more dramatic effect with SR-BI-expressing cells compared with control cells is a result of the enhanced rate of cholesterol desorption induced by SR-BI expression. Conversely, depletion of HDL PC leads to an insufficient solubilization of the desorbed molecules, thereby limiting the rate of SR-BI-mediated cholesterol efflux. The finding of a shift in FC localization from the core to the surface of PC-depleted HDL might also contribute to the ineffective solubilization of desorbed cholesterol molecules (25).

The lack of an effect on SR-BI-mediated influx of FC when HDL PC content was manipulated is consistent with studies showing that the desorption rate of FC from model membranes, and thus presumably the surface of HDL particles, is not affected by the PC to cholesterol ratio (46). The sphingomyelin effect on influx is likely due to a decrease in the rate of FC desorption from the surface of the HDL resulting from the high affinity that sphingomyelin has for cholesterol (49, 50). This would be consistent with prior studies showing that the rates of desorption from model membranes are decreased when vesicles containing sphingomyelin are compared with those containing PC alone.

Physiological Significance-- The present studies demonstrate that HDL PC and sphingomyelin content play an important role in determining the rate and direction of net transfer of FC with SR-BI-expressing cells. Phosphatidylcholine is the main phospholipid subclass present in HDL. From the current studies, it is expected that factors which modulate both HDL PC content and SR-BI expression would have profound effects on RCT. There are many studies that support the belief that HDL phospholipid plays a key role in RCT. Previous studies from this laboratory have shown that efflux to serum from hepatoma cells, which naturally express high levels of SR-BI, is correlated better to HDL phospholipid than to other serum parameters (5, 20, 21, 51). Other studies have shown that patients with coronary artery disease have lower HDL phospholipid levels compared with controls (22). In addition, patients with type II Alagille syndrome have HDL phospholipid to protein ratios similar to the PC-enriched HDL in the current studies (52). These particles are much more efficient than control HDL at stimulating efflux, and patients with this disease rarely have coronary heart disease (52).

The lipid composition of HDL differs among subclasses and results from the action of a number of factors. These factors, acting in concert, will serve to regulate the bi-directional flux of FC between SR-BI-expressing cells and HDL in both the peripheral lymph, where the process of RCT is initiated, and in the liver where the final steps of RCT occur (for reviews see Refs. 53 and 54). Among the different factors are those in serum, which include lecithin cholesterol acyltransferase, cholesteryl ester transfer protein, and phospholipid transfer protein, as well as a number of cell-surface lipases such as hepatic triglyceride lipase and lipoprotein lipase. It is also likely that the recently discovered secretory tissue phospholipases, such as the endothelially derived phospholipase (55), phospholipase A₂ (56), and sphingomyelinase (57) will alter HDL phospholipid and, consequently, influence the mobilization of FC from the periphery. Thus, the present findings suggest that tissue- or cell-localized phospholipase activity may alter net FC flux via SR-BI-mediated mechanisms by changing the phospholipid content of HDL particles in the immediate vicinity of the cell.

	ACKNOWLEDGEMENTS

We thank Vinh Nguyen, Faye Baldwin, Ruixue Wang, and Margret Nickel for the excellent technical assistance.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants HL22633, HL58012, and HL63768 and additional funding from Pfizer Central Research.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Pharmacological Sciences, University Medical Center, State University of New York, Stony Brook, NY 11794. Tel.: 631-444-3083; Fax: 631-444-3011; E-mail: Dave@pharm.sunysb.edu.

Published, JBC Papers in Press, August 29, 2000, DOI 10.1074/jbc.M006924200

	ABBREVIATIONS

The abbreviations used are: RCT, reverse cholesterol transport; BSA, bovine serum albumin; DMPC, dimyristoyl-sn-glcerophosphocholine; FC, free cholesterol; HDL, high density lipoprotein; POPC, 1-palmitoyl-2-oleoyl-sn-glycerophosphocholine; POPS, palmitoyl-2-oleoyl-sn-glycerophosphoserine; PLA₂, phospholipase A₂; SR-BI, scavenger receptor, class B, type 1; SUV, small unilamellar vesicles; MLVs, multilamellar vesicles; DMEM, Dulbecco's modified Eagle's medium; CS, calf serum.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit