High density lipoprotein prevents oxidized low density lipoprotein-induced inhibition of endothelial nitric-oxide synthase localization and activation in caveolae.

Oxidized LDL (oxLDL) depletes caveolae of cholesterol, resulting in the displacement of endothelial nitric-oxide synthase (eNOS) from caveolae and impaired eNOS activation. In the present study, we determined if the class B scavenger receptors, CD36 and SR-BI, are involved in regulating nitric-oxide synthase localization and function. We demonstrate that CD36 and SR-BI are expressed in endothelial cells, co-fractionate with caveolae, and co-immunoprecipitate with caveolin-1. Co-incubation of cells with 10 microgram/ml high density lipoprotein (HDL) prevented oxLDL-induced translocation of eNOS from caveolae and restored acetylcholine-induced nitric-oxide synthase stimulation. Acetylcholine caused eNOS activation in cells incubated with 10 microgram/ml oxLDL (10-15 thiobarbituric acid-reactive substances) and blocking antibodies to CD36, whereas cells treated with only oxLDL were unresponsive. Furthermore, CD36-blocking antibodies prevented oxLDL-induced redistribution of eNOS. SR-BI-blocking antibodies were used to demonstrate that the effects of HDL are mediate by SR-BI. HDL binding to SR-BI maintained the concentration of caveola-associated cholesterol by promoting the uptake of cholesterol esters, thereby preventing oxLDL-induced depletion of caveola cholesterol. We conclude that CD36 mediates the effects of oxLDL on caveola composition and eNOS activation. Furthermore, HDL prevents oxLDL from decreasing the capacity for eNOS activation by preserving the cholesterol concentration in caveolae and, thereby maintaining the subcellular location of eNOS.

Hypercholesterolemia-induced vascular disease and atherosclerosis are characterized by an early, selective decrease in the bioavailability of endothelium-derived nitric oxide (NO) (1). Responsiveness to receptor-dependent stimuli, such as acetylcholine, is decreased in the initial phase of the disease process, whereas responsiveness to receptor-independent stimuli such as the calcium ionophore A23187 is not altered. Therefore, the early pathogenesis is characterized by attenuated endothelial NO production in response to extracellular stimuli, even though the capacity for maximal enzyme activation and the breakdown of NO are not affected. As the disease progresses, nonspecific inhibition of NO bioavailability occurs, which is at least partly due to enhanced inactivation of NO by superoxide anions (2)(3)(4). The chronic inhibition of NO synthesis in rabbit models of hypercholesterolemia accelerates the development of vascular dysfunction and intimal lesions, providing additional evidence that the impairment of NO synthesis promotes atherogenesis (5). In vitro investigations have further shown that oxLDL 1 inhibits NO-mediated responses (6).
Numerous studies have demonstrated that the endothelial isoform of NO synthase (eNOS) is localized in plasmalemmal caveolae and that caveolin is a negative regulator of eNOS enzymatic activity (7). Caveolae are lipid domains that typically represent about 1-4% of the total plasma membrane surface area (8). The structure and function of caveolae is dependent on the amount of cholesterol associated with the domain. Cholesterol plays a key role in controlling the morphology of caveolae and the localization of proteins to caveolae (8 -11). High concentrations of cholesterol induce caveolae to invaginate, whereas low cholesterol concentrations generate a flattened morphology (8 -11). We recently demonstrated that oxLDL depletes plasma membrane caveolae of cholesterol, leading to the translocation of eNOS from caveolae to an intracellular membrane compartment and the inhibition of acetylcholine-induced eNOS activation (12).
The class B scavenger receptors, CD36 and SR-BI, are thought to be involved in the process of atherogenesis and both are enriched in caveolae (13)(14)(15). One of the striking features of CD36 and SR-BI is their broad and overlapping ligand binding specificity. For instance, both receptors bind HDL, native LDL, and oxidized LDL, but not the many polyanions that are bound by the class A scavenger receptors (16,17). The functional significance of many of these receptor-ligand interactions is not known. We recently demonstrated that oxLDL binding to CD36, but not SR-BI, promotes the efflux of caveola cholesterol to oxLDL. In addition, we have shown that caveolae are the initial sites of SR-BI-mediated selective uptake of HDL cholesterol esters (18), and that caveolae are acceptor membranes for newly synthesized cholesterol (11).
To further understand the mechanisms by which cholesterol status affects endothelial NO production, the present studies were designed to determine whether HDL modifies the effects of oxLDL on eNOS subcellular localization and function in cultured endothelial cells. Since there is a negative relationship between serum HDL levels and the risk for atherosclerosis (19,20), we tested the hypothesis that HDL reverses the negative effects of oxLDL on eNOS localization and activation in caveolae. Additional experiments were performed to determine the roles of CD36 and SR-BI receptors in these processes, and the basis for the effects of HDL.
Isolation of Caveolae-Caveolae membranes were isolated as described previously (22,23). This procedure generates a highly purified plasma membrane microdomain that is free from intracellular markers and bulk plasma membrane markers. This method has been used extensively to characterize caveola membranes (11, 22, 24 -27).
Immunoprecipitations-Protein A-Sepharose beads were first blocked by incubation for 4 h at 4°C with Chinese hamster ovary cell lysate (200 g/ml) plus 30 mg/ml bovine serum albumin in immunoprecipitation buffer (150 mM NaCl, 0.5% Triton X-100, 50 mM Tris, pH 8.0). Blocked beads were then used to preclear the experimental fractions that had been adjusted to 0.5% (v/v) Triton X-100. Precleared fractions were incubated for 18 h at 4°C with the appropriate antibody before adding blocked Protein A-Sepharose beads and incubating an additional 2 h at 4°C. The beads were collected by centrifugation, washed five times in high salt (500 mM NaCl) immunoprecipitation buffer, and then placed in Laemmli sample buffer. Immunoprecipitated proteins were detected by immunoblotting.
Electrophoresis and Immunoblots-Cellular fractions were dissolved in 0.015% (w/v) deoxycholate, concentrated by precipitation with 7% (w/v) trichloroacetic acid, and washed in acetone. Pellets were suspended in 1ϫ sample buffer plus 1.2% (v/v) ␤-mercaptoethanol and heated to 95°C for 5 min immediately prior to loading. Proteins were separated on a 12.5% polyacrylamide gel at 50 mA (constant current) and subsequently transferred to polyvinylidene difluoride membrane at 50 V (constant voltage) for 2 h. Membranes were blocked with blotting buffer for 60 min at 22°C. Primary antibodies were diluted in blotting buffer and incubated with blocked membranes for 60 min at 22°C. Membranes were washed four times for 10 min in wash buffer. Horseradish peroxidase-conjugated IgGs directed against the appropriate host IgGs were diluted and incubated with membranes as described for primary antibodies. Membranes were washed four times for 10 min in wash buffer and visualized using chemiluminescence.
Isolation, Labeling, and Oxidation of Lipoproteins-LDL (d ϭ 1.019 -1.05 g/ml) and HDL (d ϭ 1.063-1.21 g/ml) were isolated from fresh human plasma by density gradient ultracentrifugation as described previously (28,29). The HDL 3 subfraction (d ϭ 1.13-1.18 g/ml) was isolated from other HDL subfractions using a density gradient fractionator (Isco). The HDL 3 subfraction was used for all of the experimental treatments described herein. The purity of each lipoprotein fraction was assayed by SDS-PAGE and Coomassie staining. LDL apolipoproteins were iodinated by the iodine monochloride method (30) to a specific activity of 400 -600 cpm/ng LDL protein. Oxidized LDL was pre-pared by incubating fresh LDL with 10 M CuSO 4 at 37°C for 16 h. The material was dialyzed against a sterile solution (150 mM NaCl, 1 mM EDTA, 100 g/ml polymyxin B, pH 7.4) and then sterilized by filtration. The cholesterol, triglyceride (31), and protein (32) content of the lipoprotein were determined by standard methods. The homogeneity of each preparation was determined by SDS-PAGE. The extent of LDL oxidation was estimated by agarose gel electrophoresis and by measuring the amount of thiobarbituric acid-reactive substances generated with a colorimetric assay for malondialdehyde (33).
Activation of NOS in Intact Cells-NOS activation was determined in whole endothelial cells grown in 24-well plates by measuring [ 3 H]Larginine conversion to [ 3 H]L-citrulline using methods previously described. This procedure provides a direct assessment of the acute activation of existing eNOS while keeping signal transduction mechanisms intact (34). Subconfluent cells were placed in L-arginine-deficient, serum-free media containing 1% antibiotic-antimycotic mixture, 0.15% Nystatin, 0.15% gentamycin and 0.10% Tylosin for 18 h at 37°C. The medium was replaced with phosphate-buffered saline (PBS, 500 l/ well), pH 7.4, containing 120 mM NaCl, 4.2 mM KCl, 2.5 mM CaCl 2 , 1.3 mM MgSO 4 , 7.5 mM glucose, 10 mM HEPES, 1.2 mM Na 2 HPO 4 , and 0.37 mM KH 2 PO 4 for a 60-min preincubation at 37°C. In selected wells (blanks), 500 l of 1N trichloroacetic acid was added in lieu of PBS. After the preincubation period, the incubation for NOS activation was initiated by aspirating the PBS from the wells and replacing it with 400 l of PBS containing 0.75 Ci/ml [ 3 H]L-arginine. The plates were incubated at 37°C for 15 min in the absence of exogenous stimulation (basal) or in the presence of acetylcholine (10 Ϫ6 M), with or without the addition of 10 g/ml oxLDL and/or HDL. The NOS reaction was terminated by adding 500 l of ice-cold 1 N trichloroacetic acid to each well. The preincubations were performed in the presence of oxLDL, nLDL, IgGs, or buffer as described under "Results." The cells were freezefractured twice in liquid nitrogen for 2 min with thawing at 37°C for 5 min, and scraped with a rubber spatula. The contents of each well were aspirated and transferred to ice-cold sialonized glass test tubes. Ether extraction was performed three times with water-saturated ether. The samples were neutralized with 1.5 ml of 25 mM HEPES, pH 8, applied to Dowex AG50WX-8 (Tris form) columns, and eluted with 1 ml of 40 mM HEPES buffer, pH 5.5, containing 2 mM EDTA and 2 mM EGTA. Uptake Assays-The selective uptake of cholesterol esters from HDL into cells was determined using [ 3 H]cholesterol oleate (18). Confluent cells were rinsed twice with PBS (37°C). Growth medium containing 5% human lipoprotein-deficient serum (LPDS) and 10 g/ml [ 3 H]cholesterol oleate-HDL was added to the cells for the indicated times. Following incubation, uptake was terminated by aspirating the medium and washing the cell monolayers four times with Tris-saline (4°C). The cells were dissolved in 1 M NaOH and the amount of radiation determined by scintillation counting. Total cellular protein was determined by Lowry (32).
Radiolabeled Cholesterol Determination-Thin layer chromatography and liquid scintillation counting was used to measure the amount of [ 3 H]sterol in each sample (11). Each sample was adjusted to a volume of 1 ml with distilled water. Methanol (1.2 ml) containing 2% (v/v) acetic acid was added to the sample before vortexing two times, 30 s each. Chloroform (1.2 ml) was then added and the sample vortexed two times, 30 s each. The organic and aqueous phases were separated in a Beckman Clinical Centrifuge at 1200 ϫ g, 15 min at room temperature. The organic phase was dried under nitrogen and then suspended in 50 l of the solvent system (80:20:1; petroleum ether:ethyl ether:acetic acid). Pure cholesterol was dissolved in the solvent system and used as a standard (5 g/spot). Lipids were visualized by charring with sulfuric acid-ethanol and heating at 180°C for 10 min. Unlabeled cholesterol was added to each fraction to facilitate visualization. The appropriate spots were scraped and the amount of radiation quantified by liquid scintillation counting.
Cholesterol Mass Measurements-To each 0.25-ml sample, 1 ml of hexane:isopropanol (3:2) was added and incubated at room temperature for 30 min. The organic phase was saved while the aqueous phase was re-extracted. The organic phases were combined and dried to completeness with nitrogen. The sample was solubilized in 600 l of 1% Triton X-100 in chloroform, and dried to completeness with nitrogen. The sample was then solubilized in 300 l of water, and the total cholesterol was determined as described previously (35).

Expression and Localization of CD36 and SR-BI-
We have previously demonstrated that oxLDL inhibits receptor-dependent stimulation of eNOS and alters the subcellular distribution of eNOS (12). The class B scavenger receptors, CD36 and SR-BI, bind oxLDL. Therefore, to examine the possible involvement of these receptors in mediating the effect of oxLDL on eNOS distribution and function, we determined if these proteins are expressed in endothelial cells. We subfractionated quiescent endothelial cells to isolate caveolae and resolved 15 g from each subcellular fraction by SDS-PAGE. The material was transferred to nylon and immunoblotted with CD36, SR-BI, caveolin-1, clathrin, and transferrin receptor IgGs (Fig. 1A). CD36, SR-BI, and caveolin-1 were highly enriched in the caveola fraction. The yield of CD36, SR-BI, and caveolin-1 in the caveola fraction was determined by immunoblot analysis in the linear range of detection (data not shown). The estimated yield, with respect to the plasma membrane fraction, of CD36 was 56 Ϯ 9%, SR-BI was 62 Ϯ 7%, and caveolin-1 was 51 Ϯ 5%. The non-caveola proteins, clathrin and transferrin receptor, were not associated with the caveola fraction.
Recently Stan et al. (36) and Waugh et al. (37) suggested that isolated caveolae are contaminated with vesicles with similar biophysical properties as caveolae. Therefore, to verify that CD36 and SR-BI are associated with caveolin-containing membranes we used caveolin-1 IgG to immunoprecipitate caveola membranes from caveolin-enriched membrane fractions. The immunoprecipitated material (PEL) and the remaining supernatant (SUP) were then resolved by SDS-PAGE and immunoblotted for caveolin-1, CD36, and SR-BI. Caveolin-1 IgG quantitatively co-immunoprecipitated CD36 (Fig. 1B, PEL versus SUP) and SR-BI (Fig. 1C, PEL versus SUP). These findings are consistent with previous studies demonstrating the localization of CD36 (13,14) and SR-BI (15) to caveolae.
Effect of HDL on eNOS-Since plasma levels of HDL cholesterol are inversely correlated with the risk of developing atherosclerosis, we next determined if HDL reverses the effects of oxLDL on eNOS localization and function. Cells were treated with 10 g/ml lipoprotein in LPDS for 1 h at 37°C and then subfractionated to isolate caveolae. For experiments with both oxLDL and HDL, oxLDL (10 g/ml) was incubated with cells for 60 min, and then HDL (10 g/ml) was added for an additional 15 min prior to subfractionation. Equal amounts of protein (15 g) from each subcellular fraction were analyzed for the presence of eNOS and caveolin-1 by immunoblotting ( Fig.  2A). eNOS and caveolin-1 were highly enriched in caveolae isolated from cells treated with nLDL or HDL. In contrast, oxLDL caused the translocation of eNOS and caveolin-1 from caveolae to an intracellular membrane fraction. The addition of HDL to the treatment medium containing oxLDL induced the redistribution of eNOS and caveolin-1 back to the caveola fraction. Furthermore, HDL restored acetylcholine-induced stimulation of eNOS activity (Fig. 2B).
CD36 Mediates the Effects of oxLDL on eNOS-We used a commercially available CD36-blocking antibody (FA6.152) to determine if CD36 mediates the effects of oxLDL on eNOS location and function. Cells were treated for 1 h at 37°C with 10 g/ml oxLDL in the presence of LPDS, LPDS plus nonspecific IgG (2 g/ml), or LPDS plus CD36 IgG (2 g/ml). The cells were then subfractionated to isolate caveolae, and 15 g of protein from each fraction was analyzed by immunoblot for the presence of eNOS and caveolin-1 (Fig. 3A). Incubation of cells with nonspecific IgG did not prevent the oxLDL-induced translocation of caveolin-1 and eNOS to the intracellular membrane fraction. However, co-incubation with CD36-blocking antibody inhibited oxLDL-induced movement of caveolin-1 and eNOS. We also determined if the CD36-blocking antibody would prevent oxLDL-induced suppression of acetylcholine-induced eNOS stimulation (Fig. 3B). Consistent with the effects on eNOS localization, nonspecific IgG did not prevent oxLDL-induced inhibition of eNOS activity whereas CD36 IgG permitted eNOS activation.
SR-BI Mediates the Effects of HDL-We used a SR-BI-blocking antibody characterized by Temel et al. (38) to determine if HDL binding to SR-BI is responsible for reversing the effects of oxLDL on eNOS distribution and function. Cells were treated with oxLDL or oxLDL plus HDL as described above in the presence of nonspecific IgG (500 g/ml) or SR-BI IgG (500 g/ml). The cells were then subfractionated to isolate caveolae, and 15 g of protein from each fraction was analyzed by immunoblot for the presence of eNOS and caveolin-1 (Fig. 4A). In the presence of nonspecific IgG, HDL was able to restore the localization of eNOS and caveolin-1 to caveolae. However, coincubation with SR-BI IgG, which blocks the binding of HDL to SR-BI (38), prevented HDL from preserving the localization of caveolin-1 and eNOS to caveolae. In addition, SR-BI-blocking IgG interfered with HDL restoration of acetylcholine-induced stimulation of eNOS (Fig. 4B). 1. CD36 and SR-BI are co-purified and co-immunoprecipitated with caveolin-1. Endothelial cells were processed to isolate caveolae by a standard procedure (22,23). A, an equal amount of protein (15 g) from each fraction was resolved by SDS-PAGE and immunoblotted with IgGs for CD36, SR-BI, caveolin-1, clathrin, and transferrin receptor (TR). The immunoblots were developed by the method of chemiluminescence. The caveolin, SR-BI, and CD36 immunoblots were exposed for 30 s, and the clathrin and transferrin receptor immunoblots were exposed for 2 min. PNS, postnuclear supernatant; CYTO, cytosol; IM, total intracellular membranes; PM, plasma membrane; CM, caveola membrane. B and C, IgG directed against caveolin-1 (2 g/ml) was used to immunoprecipitate caveolae from the caveola-enriched subcellular fraction. The entire pellet (PEL) and the entire supernatant (SUP) were resolved by SDS-PAGE and immunoblotted for CD36 and caveolin-1 (B) or SR-BI and caveolin-1 (C). The immunoblots were developed by the method of chemiluminescence (45-s exposures). Figure shows representative data from three independent experiments.
HDL Maintains the Sterol Content of Caveolae-To determine if HDL prevents oxLDL-induced depletion of caveola cholesterol, we incubated cells with 10 g/ml oxLDL, 10 g/ml HDL, or both for 0 -60 min, and then measured the mass of cholesterol associated with caveolae. Fig. 5A shows that oxLDL depleted caveola-associated cholesterol by 40 min, whereas treatment with 10 g/ml HDL did not alter the total cholesterol mass associated with caveolae (Fig. 5B). The addition of 10 g/ml HDL to the oxLDL treatment medium attenuated the capacity of oxLDL to deplete caveolae of cholesterol (Fig. 5C).
We next determined if HDL-mediated protection from cholesterol depletion by oxLDL was due to the inhibition of cholesterol transport out of caveolae by oxLDL, or the transport of cholesterol into caveolae from HDL. Cells were radiolabeled with [ 14 C]acetate for 18 h to label intracellular cholesterol (22). HDL and oxLDL (10 g/ml each) were added simultaneously FIG. 2. HDL restores the localization and activation of eNOS in caveolae. A, endothelial cells were treated for 1 h at 37°C with 10 g/ml nLDL, HDL, or oxLDL and subfractionated to isolate caveolae. For experiments with both oxLDL and HDL, oxLDL (10 g/ml) was incubated with cells for 60 min, and then HDL (10 g/ml) was added for an additional 15 min prior to subfractionation. Equal amounts of protein (15 g) from each subcellular fraction was analyzed for eNOS and caveolin-1 by immunoblotting. The immunoblots were developed by the method of chemiluminescence (60 s for eNOS and 20 s for caveolin-1). PNS, postnuclear supernatant; CYTO, cytosol; IM, total intracellular membranes; PM, plasma membrane; CM, caveola membrane. Representative data from five independent experiments is shown. B, after treating endothelial cells as described above, the incubation for NOS activity was initiated by adding 0.75 Ci/ml [ 3 H]L-arginine to the wells. The plates were incubated at 37°C for 15 min in the absence of exogenous stimulation (Basal) or in the presence of acetylcholine (Ach, 10 Ϫ6 M). The [ 3 H]L-citrulline formed was quantified by liquid scintillation spectroscopy. Findings were confirmed in four independent experiments, mean Ϯ S.E., n ϭ 4. *, p Ͻ 0.05 versus basal.

FIG. 3. CD36-blocking antibodies prevent oxLDL-induced redistribution of eNOS and inhibition of acetylcholine-stimulated eNOS activity.
A, endothelial cells were treated for 1 h at 37°C with 10 g/ml oxLDL in the presence of LPDS, LPDS plus nonspecific (NS) IgG (2 g/ml) or LPDS plus CD36 IgG (FA6.152) (2 g/ml). The cells were then subfractionated to isolate caveolae, and 15 g of protein from each fraction was analyzed by immunoblot for eNOS and caveolin-1. The immunoblots were developed by the method of chemiluminescence (60 s for eNOS and 20 s for caveolin-1). PNS, postnuclear supernatant; CYTO, cytosol; IM, total intracellular membranes; PM, plasma membrane; CM, caveola membrane. Representative data from three independent experiments is shown. B, subconfluent endothelial cells were placed in phosphate-buffered saline containing either nLDL or oxLDL at a final concentration of 10 g/ml, and either nonspecific IgG (2 g/ml) or CD36 IgG (FA6.152) (2 g/ml) for a 60-min preincubation at 37°C. After the preincubation period, the incubation for NOS activity was initiated by adding 0.75 Ci/ml [ 3 H]L-arginine to the wells. The plates were incubated at 37°C for 15 min in the absence of exogenous stimulation (Basal) or in the presence of acetylcholine (Ach, 10 Ϫ6 M). The [ 3 H]L-citrulline formed was quantified by liquid scintillation spectroscopy. Findings were confirmed in four independent experiments, mean Ϯ S.E., n ϭ 4. *, p Ͻ 0.05 versus basal.
for 0 -60 min. The cells were then processed to measure the amount of [ 14 C]cholesterol associated with caveolae and the mass of cholesterol associated with caveolae. The amount of [ 14 C]cholesterol associated with caveolae decreased rapidly and was depleted by 40 min (Fig. 5D). However, the mass of cholesterol associated with caveolae was not changed (Fig. 5D).
These data suggest that HDL does not inhibit cholesterol depletion but rather serves as a cholesterol donor. To test this, cellular cholesterol pools were radiolabeled as described above and HDL was labeled with [ 3 H]cholesterol ester (18). [ 3 H]HDL and oxLDL were added to cells as described above. Fig. 5E

DISCUSSION
The present observations reveal that CD36 mediates the rapid and dramatic effects of oxLDL on eNOS localization and function FIG. 4. SR-BI-blocking antibodies prevent the effects of HDL on eNOS localization and activation in caveolae. A, endothelial cells were treated for 1 h at 37°C with 10 g/ml oxLDL, or oxLDL and 10 g/ml HDL in the presence of LPDS plus nonspecific IgG (500 g/ml) or LPDS plus SR-BI IgG (500 g/ml). The cells were then subfractionated to isolate caveolae, and 15 g of protein from each fraction was analyzed by immunoblot for eNOS and caveolin-1. The immunoblots were developed by the method of chemiluminescence (60 s for eNOS and 20 s for caveolin-1). PNS, postnuclear supernatant; CYTO, cytosol; IM, total intracellular membranes; PM, plasma membrane; CM, caveola membrane. Representative data from four independent experiments is shown. B, subconfluent endothelial cells were treated as described above with the exception that LPDS was replaced with phosphatebuffered saline. After the preincubation period, the incubation for NOS activity was initiated by adding 0.75 Ci/ml in caveolae. In addition, we demonstrate that SR-BI mediates the ability of HDL to protect caveolae from oxLDL-induced depletion of cholesterol and eNOS redistribution. These findings indicate that at least some of the athero-protective features of HDL may be due to direct effects on endothelial cell signal transduction mechanisms that are localized to plasmalemmal caveolae.
The class B scavenger receptors, CD36 and SR-BI, were expressed in the endothelial cells used in this investigation (Fig. 1). Subcellular fractionation studies suggested that these proteins are associated with caveolae. Recently, Stan et al. (36) and Waugh et al. (37) have used co-immunoprecipitation experiments to suggest that isolated caveolae are contaminated with non-caveola low density membranes. To confirm that CD36 and SR-BI are associated with caveola membranes, we first isolated caveolae and then immunoprecipitated the membrane fragments with caveolin-1 IgG. Greater than 90% of CD36 and SR-BI co-immunoprecipitated with caveolin-1 from isolated caveola fractions. These new data along with previously published confocal studies (15) demonstrate that a significant proportion of CD36 and SR-BI are associated with endothelial caveolae.
Because CD36 is known to bind oxLDL, we determined if this protein is responsible for the effects of oxLDL on caveolae and eNOS function. We demonstrated that the addition of a CD36blocking antibody to the oxLDL treatment medium inhibits the ability of oxLDL to cause the redistribution of eNOS and caveolin-1 (Fig. 3A). Furthermore, the addition of CD36-blocking antibodies maintained the ability of eNOS to be activated by acetylcholine (Fig. 3B). HDL protected caveolae from the effects of oxLDL; however, SR-BI-blocking antibodies eliminated this protection (Fig. 4). SR-BI has been reported to bind oxLDL but SR-BI-blocking antibodies did not affect the oxLDL-induced reduction in eNOS activity (data not shown). These data provide a mechanism whereby oxLDL and HDL can specifically interact with caveolae, namely through the scavenger receptors CD36 and SR-BI.
We further examined the mechanism of HDL protection. Two possibilities were tested. First, does HDL prevent cholesterol efflux to oxLDL? Our data demonstrate that HDL does not prevent the efflux of caveola sterols to oxLDL (Fig. 5). Second, we tested the possibility that HDL serves as a sterol donor and replaces the sterol effluxed to oxLDL. By radiolabeling the cellular sterol pool, we were able to demonstrate that the radiolabeled endogenous sterol in caveolae was depleted in the presence of oxLDL and HDL without a corresponding decrease in caveola cholesterol mass. These data suggested that HDL replaces the sterol that is lost from caveolae upon exposure to oxLDL. We previously demonstrated that SR-BI mediates the selective uptake of cholesterol esters into caveolae (18). To directly determine if HDL replaces caveola sterols, we radiolabeled cellular cholesterol pools with 14 C and HDL with [ 3 H]cholesterol ester. The removal of endogenous [ 14 C]sterol from caveolae was matched by an accumulation of [ 3 H]cholesterol ester in caveolae. These data demonstrate that the lipid composition of caveolae can be regulated by CD36, SR-BI, oxLDL, and HDL. Importantly, the lipoprotein-mediated alterations in caveola composition are directly related to a physiologic phenomena, namely the generation of nitric oxide.
In summary, we have shown that oxLDL-induced redistribution of eNOS from caveolae and the subsequent inability to efficiently activate eNOS with acetylcholine is mediated by the class B scavenger receptor, CD36. These findings provide a potential mechanism by which elevations in oxLDL rapidly attenuate the capacity for NO production by the endothelium. As such, they may at least partially explain the repeated observation in many experimental models and in humans that hypercholesterolemia-induced vascular disease and atheroscle-rosis are characterized by an early, selective impairment of agonist-stimulated endothelium-dependent relaxation, and that this is due to a decrease in bioavailable endotheliumderived NO (1). Furthermore, we have demonstrated that HDL prevents the depletion of caveola-cholesterol by oxLDL in a SR-BI-dependent manner. This process may be critical to the athero-protective effects of HDL. Additional studies of the effects of oxLDL and HDL on caveolae composition and eNOS activation will increase our fundamental knowledge of the regulation of endothelial NO production under both normal conditions and during altered cholesterol balance, thereby further revealing the pathogenic link between hypercholesterolemia and endothelial dysfunction.