A novel family of atherogenic oxidized phospholipids promotes macrophage foam cell formation via the scavenger receptor CD36 and is enriched in atherosclerotic lesions.

The macrophage scavenger receptor CD36 plays an important role in binding and uptake of oxidized forms of low-density lipoprotein (LDL), foam cell formation, and lesion development during atherosclerosis. The structural basis of CD36-lipoprotein ligand recognition is an area of intense interest. In a companion article we reported the characterization of a structurally conserved family of oxidized choline glycerophospholipids (oxPC(CD36)) that serve as novel high affinity ligands for cells stably transfected with CD36, mediating recognition of multiple oxidized forms of LDL (Podrez, E. A., Poliakov, E., Shen, Z., Zhang, R., Deng, Y., Sun, M., Finton, P., Shan, L., Gugiu, B., Fox, P. L., Hoff, H. F., Salomon, R. G., and Hazen, S. L. (July 8, 2002) J. Biol. Chem. 277, 10.1074/jbc.M203318200). Here we use macrophages from wild-type and CD36 null mice to demonstrate that CD36 is the major receptor on macrophages mediating recognition of oxPC(CD36) species when presented (+/- plasma) in pure form, within PC bilayers in small unilamellar vesicles, and within liposomes generated from lipid extracts of native LDL. We also show that oxPC(CD36) promote CD36-dependent recognition when present at only a few molecules per particle, resulting in macrophage binding, uptake, metabolism, cholesterol accumulation, and foam cell formation. Finally, using high performance liquid chromatography with on-line electrospray ionization tandem mass spectrometry (LC/ESI/MS/MS), we demonstrate that oxPC(CD36) are generated in vivo and are enriched in atherosclerotic lesions. Collectively, our data suggest that formation of this novel family of oxidized phospholipids participates in CD36-mediated recognition of oxidized lipoproteins and foam cell formation in vivo.

CD36 is a multifunctional cellular receptor with broad ligand specificity (1,2). It is expressed in a number of cells including microvascular endothelial cells, platelets, adipocytes, striated muscles, macrophages, and some vascular smooth muscle cells (1,2). CD36 regulates cellular adhesion and angiogenesis, serving as a receptor for thrombospondin; it also serves as a scavenger receptor in macrophages, mediating uptake of apoptotic cells and modified lipoproteins, and participates in carbohydrate and lipid metabolism, modulating insulin resistance and long chain fatty acid transport (3)(4)(5)(6)(7)(8). CD36 has recently been implicated in a variety of pathologic conditions, including atherosclerosis, diabetes, and cardiomyopathy. Perhaps the most compelling data on the role of CD36 in atherosclerosis are from studies of CD36-deficient mice, which show a 70 -80% reduction in aortic lesion size (9). In vitro experiments demonstrate that macrophages from CD36deficient mice take up different forms of oxidized LDL 1 poorly and are resistant to foam cell formation, providing a mechanism for the atheroprotection observed in CD36 null mice (9,10). Given the potential clinical significance of this receptor, it is important to know the nature of the ligand(s) in oxidized LDL (oxLDL) that are recognized by CD36 and support foam cell formation in vivo. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ Recipient of a Jane Coffin Childs Memorial Fund for Medical Research fellowship.
The scavenger receptor functions of CD36 toward oxLDL may, in theory, result from recognition of the lipid moiety, apoB-100 or apoB-100 modified by oxidized lipids. Oxidized lipids were first suggested to participate in macrophage recognition of oxLDL based on studies using liposomes generated from lipid extracts of LDL oxidized by free Cu 2ϩ (Cu 2ϩ -oxLDL) and mouse peritoneal macrophages (MPM) (11). A role for CD36 as the macrophage receptor responsible for recognition of oxidized lipids in Cu 2ϩ -oxLDL has subsequently been shown (10,12). Oxidized phospholipids covalently linked to apolipoprotein B-100 (apoB) in extensively oxidized LDL (e.g. Cu 2ϩ -oxLDL) have been suggested to serve as ligands for CD36, based on competition studies using reconstituted apoB-100 from Cu 2ϩ -oxLDL and studies employing monoclonal antibodies to phospholipid-protein adducts (12,13).
We recently employed a combination of mass spectrometry and both analytical and synthetic chemistry to isolate and structurally define a novel family of oxidized choline glycerophospholipids (oxPC) that, in an unbound state, serve as high affinity ligands for CD36 (14). We further demonstrated their formation during the oxidization of LDL by multiple distinct pathways and their participation in CD36-mediated recognition of different forms of oxidized LDL (14). The highly conserved nature of the critical structural elements required for oxidized phospholipids to serve as ligands for CD36 were defined, namely, a phospholipid with a truncated sn-2 acyl group that incorporates a terminal ␥-hydroxy(or oxo)-␣,␤-unsaturated carbonyl (oxPC CD36 ) (Scheme 1) (14). Finally, in competition studies using cells transfected with CD36, we demonstrated that the binding site for oxPC CD36 species (and oxLDL) on CD36 is spatially and functionally distinct from that of other known CD36 ligands (14). Collectively, these results suggested that formation of this novel family of oxidized phospholipids might play an important role in CD36-mediated recognition of oxidized lipoproteins, senescent or apoptotic cells, and foam cell formation. However, the physiological relevance of oxPC CD36 has not yet been established. Namely, their formation in vivo, their ability to promote cholesterol loading and foam cell formation via CD36, and the levels of oxPC CD36 required to promote CD36-mediated recognition by macrophages have not yet been explored.
In this study we demonstrate that oxPC CD36 inserted into small unilamellar phospholipid vesicles or liposomes containing neutral lipids derived from LDL can function as ligands for CD36 on macrophages, resulting in the internalization of the entire particle and subsequent cholesterol and cholesteryl ester loading. Specificity of oxPC CD36 -mediated foam cell formation in macrophages is confirmed using MPM from wild-type and CD36 null mice. Finally, a potential atherogenic role for ox-PC CD36 species in vivo is supported by our demonstration that oxPC CD36 are both enriched in atherosclerotic lesions and capable of promoting scavenger receptor recognition when present at only a few molecules per particle.

Methods
General Procedures-Human myeloperoxidase (donor: hydrogen peroxide, oxidoreductase; EC 1.11.1.7) and LDL were isolated and quantified as described (15). All buffers were treated with Chelex-100 resin (Bio-Rad) and supplemented with diethylenetriaminepentaacetic acid (DTPA) to remove trace levels of transition metal ions that might catalyze lipid oxidation during incubations. LDL was labeled with Na 125 I to a specific activity between 100 and 250 dpm/ng of protein, as described (16). Incorporation of [ 14 C]oleate into cholesteryl esters by cells following incubation with the indicated lipoproteins (50 g/ml) were determined as described (15).
Synthesis of Phospholipids-Total syntheses and purification of the ␥-hydroxy-␣,␤-unsaturated aldehydic phospholipids ligands for CD36 were performed as described elsewhere (17,18). The structures of all synthetic lipids were confirmed by multinuclear NMR and high resolution mass spectrometry prior to use (17)(18)(19). Synthetic lipids were routinely analyzed by HPLC with on-line electrospray ionization tandem mass spectrometry (LC/ESI/MS/MS) as described (14). If lipids were found to be less than 98% pure, they were re-isolated prior to use.
Vesicle Preparation and Modification-Stock solutions (2 mg/ml) of small unilamellar vesicles composed of PLPC, POPC, or PAPC with varying mol % of specific oxidized phospholipids were prepared in argon-sparged sodium phosphate buffer by extrusion (10ϫ) through a 0.1-m polycarbonate filter using an Avanti Mini-Extruder Set (Avanti Polar Lipids) at 37°C. For direct binding experiments, [ 3 H]DPPC (25 Ci/mg of phospholipids) or 1 mol % of the fluorescent dye DiI was added to phospholipids.
Cells-Thioglycollate-elicited MPM from wild-type (C57BL/6) and CD36 null mice (6) were isolated and cultured as described (15). CD36 null mice were back-crossed at least six generations on a C57BL/6 background. Foam cell formation was assessed following incubation of MPM with liposomes and quantified histologically as described (20). Briefly, liposomes were formed from native LDL lipids that were extracted according to the method of Bligh and Dyer (21), dried under the nitrogen, reconstituted in argon-sparged PBS at 37°C in sealed reaction vials (under argon atmosphere) at a final concentration of 0.5 mg phospholipids/ml in the presence versus absence of the indicated oxPC species. Liposomes were generated by brief sonication and subsequent extrusion (10 passages) through an 0.1-m polycarbonate filter under an argon atmosphere. Thioglycollate-elicited MPM were incubated for 48 h in RPMI 1640 containing 1% bovine serum albumin and butylated hydroxytoluene (BHT, 20 M) in the presence of the 50 g/ml liposomes. Cytochalasin D (0.5 g/ml) was included in the medium to suppress nonspecific phagocytosis. Medium containing the appropriate modified liposomes was exchanged every 24 h of incubation. Cells were fixed with 4% formaldehyde, stained with Oil Red-O to detect intracellular neutral lipids, and counter-stained with hematoxylin. Cells containing more then 10 lipid droplets were scored as "foam cells" (20). At least 10 fields and 500 cells/condition were counted.
Phospholipid Separation and Mass Spectrometric Analysis-Lipids were initially extracted three times by the method of Bligh and Dyer (21) from tissue homogenates (generated using a ratio of 1 g of tissue:10 ml of 50 mM sodium phosphate (pH 7.0) supplemented with DTPA (2 mM) and BHT (100 M) under argon atmosphere). The combined extracts were dried rapidly under nitrogen, resuspended in methanol:H 2 O (98:2, v:v), a known amount of dimyristoylphosphatidylcholine internal standard was added, and then neutral lipids in the lipid extracts were removed by passage through a C18 minicolumn (Supelclean LC-18 SPE tubes, 3 ml; Supelco Inc., Bellefonte, PA). The recovered polar phospholipids were dried under nitrogen and stored in the dark under an argon atmosphere at Ϫ80°C until analysis within 24 h. Quantification of the various oxPC species was performed using LC/ESI/MS/MS as described recently (14). Briefly, lipids were maintained under inert atmosphere (argon or nitrogen) at all times. Mass spectrometric analyses were performed on a Quatro II triple-quadrupole mass spectrometer (Micromass, Inc., Altrincham, UK) equipped with an electrospray ionization probe and interfaced with an HP 1100 HPLC (Hewlett-Packard, Wilmington, DE). Lipids were quantified on a Prodigy ODS-18 column (250 ϫ 2 mm, 5 m, Phenomenex, Torrance, CA) at a flow rate of 0.2 ml/min. Mass spectrometric analyses were performed on-line using electrospray ionization tandem mass spectrometry in the positive ion mode with multiple reaction monitoring (MRM) mode (cone potential 60 eV/collision energy 20 -25 eV). The MRM transitions used to detect the oxidized phosphopholipids present in each fraction were the mass to charge ratio (m/z) for the molecular cation [MH] ϩ and the daughter ion m/z 184, the phosphocholine group (i.e. [MH] ϩ 3 m/z 184). A discontinuous gradient was used by mixing solvent A (0.2% formic acid in HPLC-grade water) with solvent B (0.2% formic acid in HPLC-grade methanol), as follows: initial equilibration of column at 85% solvent B; increasing from 85 to 88% solvent B from 0 -12 min; increasing to 100% solvent B from 12 to 14 min; and then isocratic elution with solvent B from 14 to 20 min. Distinct oxidized phospholipid species were identified by using m/z for protonated parent ion 3 daughter ion transitions specific for each individual phospholipid and their retention times, as described earlier (14). OV-PC and ND-PC were quantified similarly but also by monitoring at the m/z for the transition between the hemiacetal formed with methanol for each analyte and the loss of polar head group (m/z 184) as described (14). Calibration curves constructed with a fixed amount of dimyristoylphosphatidylcholine and varying mol % of each synthetic oxidized PC species were used to correct for the differences in ionization response factors observed among the different lipids. In additional preliminary studies the quantification methods employed were validated independently for each analyte by demonstrating results identical to those obtained by the method of standard additions. In independent studies the overall recovery of each synthetic lipid (when analyzed at the trace levels observed in biological samples with POPC carrier) was confirmed to be in excess of 80% under the conditions employed.
Tissue Storage and Processing-The thoracic aorta from Watanabe heritable hyperlipidemic (WHHL) rabbits was isolated, rinsed in argonsparged PBS supplemented with 100 M BHT and 100 M DTPA, submerged in the same buffer, covered in argon, flash-frozen in liquid nitrogen, and then stored at Ϫ80°C until analysis. Aortas relatively free of lipid lesions were obtained from WHHL rabbits age 10 -12 weeks, whereas aortas with confluent lesions were recovered from WHHL rabbits Ͼ6 months old. Frozen sections of aortas were pulverized with a stainless steel mortar and pestle under liquid nitrogen, the powder transferred to glass threaded test tubes equipped with PTFElined caps, and then lipids were extracted by the method of Bligh and Dyer (21) under argon in the presence of BHT (100 M). At least nine samples of aortic tissues (atherosclerosis-laden and normal) were dissected from thoracic and abdominal aortas from three animals in each group for analyses. Quantification of lipids was then performed by LC/ESI/MS/MS as described above.
Fluorescence Microscopy-Cells were cultured in Lab-Tek II ® glass chamber slides (Nalge Nunc Int., Naperville, IL), incubated in the appropriate medium in the presence of DiI-labeled PAPC vesicles containing 1-25 mol % of the indicated specific oxidized phospholipid for 2 h at 4°C (for binding studies) or for 1 h at 37°C (for uptake studies). A forty-fold excess of unlabeled POPC vesicles was included to inhibit nonspecific binding and uptake. Cells were washed three times with PBS, then fixed in 4% formaldehyde in PBS (for 1 h), washed in PBS, and mounted in Vectashield ® mounting medium with 4Ј,6-diamidino-2phenylindole (Vector Laboratories, Burlingame, CA). Fluorescent preparations were viewed with a Leica fluorescent microscope with a 50ϫ objective. Images were acquired using the Magnafire imaging system (Optronics, Goleta, CA) and further analyzed using Adobe Photoshop 5.5.
Statistics-Data represent the mean Ϯ S.D. of the indicated number of samples. Statistical analyses were made using a paired Student's t test. For all hypotheses, the significance level was 0.05.

CD36 Is the Major Receptor on Macrophages That
Accounts for the Binding and Uptake of oxPC CD36 -We have recently characterized oxidized choline glycerophospholipids with an sn-2 acyl group that incorporates a terminal ␥-hydroxy(or oxo)-␣,␤-unsaturated carbonyl as high affinity ligands for the scavenger receptor CD36 (14). Their structures, chemical names and abbreviations are shown in Scheme 1. Several receptors are present on macrophages that can recognize phospholipid ligands, including CD36, SR-BI, SR-PSOX, and others (22)(23)(24). To define both the contribution of CD36 on macrophages in the recognition of oxPC CD36 and the specificity of these oxPC species for CD36, we examined the binding of oxPC CD36 to MPM from mice homozygous for a null mutation in CD36 (CD36-KO) or from wild-type (WT) mice possessing congenic backgrounds (6). Small unilamellar vesicles were made containing PAPC as a carrier lipid, synthetic oxPC CD36 species, and trace levels of the fluorescent lipid DiI. The binding (4°C, Fig. 1a) and uptake (37°C, Fig. 1b) of the phospholipid vesicles to MPM was then examined using fluorescence microscopy. Vesicles made of only PAPC (and fluorescent-label tracer) failed to bind specifically to CD36 on MPM. Only weak (and comparable) diffuse staining was detected for both WT and CD36-KO MPM. In contrast, vesicles containing either PAPC oxidized by the myeloperoxidase (MPO)-H 2 O 2 -NO 2 Ϫ system or those containing one of the identified PC species that serve as a CD36 ligand (e.g. data for KOOA-PC shown) were readily bound and taken up by WT macrophages (Fig. 1). In contrast, binding and uptake of these vesicles by CD36-KO MPM was negligible.

Addition of oxPC CD36 to Cholesterol-laden Particles Promotes CD36-dependent Accumulation of Cholesterol and Foam Cell
Formation in Macrophages-To directly assess the functional consequences of oxPC CD36 -dependent acquisition of CD36 recognition and uptake, we tested whether addition of these ligands to cholesterol-containing vesicles or liposomes facilitates cholesterol accumulation and foam cell formation in macrophages. OxPC CD36 were incorporated into either phospholipid: cholesterol (80:20 mol:mol) vesicles (Fig. 2, a-c) or liposomes derived from lipid extracts of native LDL (Fig. 2d) and then incubated with WT and CD36-KO MPM. For these studies, trace levels (Ͻ Ͻ1 mol %) of different radioactive lipids were incorporated into the phospholipid:cholesterol vesicles to monitor uptake from distinct portions of the particles (e.g.  Fig. 2c). Incorporation of oxPC CD36 into the phospholipid:cholesterol vesicles resulted in significant CD36-dependent uptake of the particles by MPM (Fig. 2,  a-c). Qualitatively similar results were observed regardless of which radiolabeled tracer was used, consistent with holoparticle uptake via CD36. Significant increases in cellular free cholesterol content, as detected by [ 14 C]cholesterol accumulation, was observed in WT but not CD36-KO macrophages after a 2-h incubation with vesicles containing ox-PC CD36 (Fig. 2b) (Fig. 2c), indicating that lipid vesicles were internalized and cholesterol esters hydrolyzed intracellularly. Similar increases in CD36-specific uptake (37°C) of ox-PC CD36 -containing particles were observed at lower mol % of oxPC CD36 species (data not shown). In parallel experiments, we compared the ability of LDL modified by the MPO-H 2 O 2 -NO 2 Ϫ system (NO 2 -LDL, 50 g/ml) to stimulate cholesterol esterification in WT MPM in the absence or presence of SCHEME 1. a, the core structural motifs conserved among the various isolated oxidized PC species that support CD36 binding activity are shown (adapted from Ref. 14). Products generated from oxidized PAPC and PLPC are similar except for the chain length of the truncated oxidized fatty acid esterified to the sn-2 position of lysophosphatidylcholine. R 1 , fatty acid such as palmitate or oleate. n, chain length (number of methylene groups). b, the structures, chemical names, and abbreviations of identified phospholipid species with CD36 binding activity derived from oxidized PAPC and PLPC are shown.
individual synthetic oxPC CD36 species (20 M). Although NO 2 -LDL markedly stimulated [ 14 C]oleate incorporation into cellular cholesterol ester pools in WT MPM in the absence of competitors, as previously reported (15), addition of the ox-PC CD36 species, but not the precursor lipids (PAPC or PLPC), inhibited [ 14 C]oleate incorporation to a level comparable with that observed following exposure to native LDL (data not shown). Finally, oxPC CD36 -dependent formation of foam cells was observed in WT but not CD36-KO MPM following a 48-h incubation with liposomes generated from lipid extracts of native LDL lipids and 5 mol % of oxPC CD36 (Fig. 2d).

OxPC CD36 Species Are Enriched in Atherosclerotic Lesions-
To determine whether oxPC CD36 species are formed in vivo, normal and atherosclerotic aortas were isolated from WHHL rabbits, and the levels of multiple distinct specific oxidized phospholipids were determined using LC/ESI/MS/MS analyses as described under "Experimental Procedures." Each of the oxPC species monitored was present in vivo. Identification of each species was based upon the detection of ions with massto-charge (m/z) ratio identical to that of the parent lipid, which following collision-induced dissociation, subsequently also gave rise to an anticipated characteristic daughter ion, all at the retention time of the appropriate authentic synthetic oxPC species. Exemplary LC/ESI/MS/MS data confirming in vivo detection of each oxPAPC species in atherosclerotic rabbit aortas are presented in Fig. 3. For illustrative purposes, the upper tracing of each panel of Fig. 3 demonstrates LC/ESI/MS/MS analysis in the positive ion mode using MRM between the transition from the protonated parent ion to the characteristic daughter ion generated by loss of the polar head group (i.e.  (Fig. 4). Interestingly, levels of oxPC CD36 species derived from PAPC oxidation were comparable with that observed for the 5-oxovaleric acid and glutaric acid esters of 2-lysoPC (OV-PC and G-PC, Fig. 4a), bioactive phospholipids initially described for their ability to promote monocyte adhe-sion to endothelial cells (19). We recently showed that OV-PC and G-PC are formed during further oxidation of PAPC-derived oxPC CD36 (14). Atherosclerotic tissue levels of oxPC CD36 species formed during PLPC oxidation (HODA-PC, KODA-PC, HDdiA-PC, and KDdiA-PC) were also increased over that of normal aorta but at lower levels than observed for the more highly oxidized ON-PC and ND-PC, the 9-oxononanoic acid and nonanedioic acid monoesters of 2-lysoPC (Fig. 4b). ON-PC and ND-PC are formed by further oxidation of the PLPC-derived oxPC CD36 and are the PLPC-derived analogs of OV-PC and G-PC (14).
Trace Levels of oxPC CD36 within Particles Are Sufficient to Confer CD36-dependent Recognition by Macrophages-The demonstrated enrichment of oxPC CD36 and their oxidation products (OV-PC, G-PC, ON-PC, and ND-PC) in atherosclerotic lesions strongly suggests that these species may participate in recognition of oxidized LDL and cellular membranes in vivo. To further assess the potential pathophysiological relevance of oxPC CD36 formation in vivo, we examined the levels of ox-PC CD36 required to confer detectable increases in CD36-specific recognition of particles using MPM from WT and CD36-KO mice. Small unilamellar vesicles (composed of PAPC) and liposomes (generated from lipid extracts of native LDL) were prepared with tracer levels (Ͻ0.1 mol %) of radiolabel (either [ 3 H]cholesterol ester or [ 14 C]DPPC) and low mol % (relative to phospholipid content) oxPC CD36 . Remarkably, whether presented within a bilayer of a vesicle or the surface of a liposome, significant (p Ͻ 0.01) and reproducible increases in CD36-specific binding of oxPC CD36 -containing particles were observed at only 0.3 mol % of oxPC CD36 species (Fig. 5, a and b). When oxPC CD36 levels were varied within liposomes, dose-dependent increases in CD36-specific binding were observed, whether monitored using a  (14). The binding of liposomes and vesicles to macrophages was then determined using WT (solid bars) and CD36-KO (open bars) MPM. c, the binding of liposomes to WT and CD36-KO MPM was performed using methods similar to those described for panel a except for the use of liposomes containing only a single species of oxPC CD36 (HDdiA-PC) at the indicated mol %. d, liposomes were generated from lipid extracts of native LDL supplemented with 1 mol % diI and either no addition (LDL Lipids) or 1 mol % (total) oxPC CD36 mixture as in panels a and b. Binding to WT and CD36-KO MPM was then monitored by fluorescence microscopy. radiolabel tracer (Fig. 5c, [ 3 H]cholesterol) or fluorophore (DiI). Fig. 5d illustrates CD36-specific binding of DiI-labeled liposomes to WT, but not to CD36-KO MPM, in the presence, but not the absence, of 1 mol % oxPC CD36 . DISCUSSION Numerous lines of evidence support a physiological role for oxPC CD36 species. First, demonstration of their enrichment in atherosclerotic lesions by LC/ESI/MS/MS analyses (Fig. 4) confirms their formation in vivo. Second, addition of oxPC CD36 to cholesterol-containing particles was shown to confer macrophage CD36-dependent binding, uptake, and metabolism of cholesterol esters with a resultant accumulation of cholesterol and foam cell formation (Fig. 2). Third, a mechanistic role for CD36 as the major scavenger receptor recognizing this novel family of oxidized choline glycerophospholipids is confirmed through studies employing WT and CD36 null mice. Fourth, further support of a physiological role of oxPC CD36 is the demonstration that incorporation of only trace levels of oxPC CD36 (as low as 0.3 mol %) conferred CD36-dependent binding to macrophages (Fig. 5). A native LDL particle possesses ϳ700 molecules of phospholipid, composed mostly of PAPC and PLPC (25). Thus, 0.3 mol % would correspond to only a few molecules of oxPC CD36 when extrapolated to an LDL particle. The levels of oxPC CD36 observed within LDL oxidized by multiple distinct oxidation systems in vitro are in excess of this level (1-5 mol %) when CD36-dependent lipoprotein recognition is observed (14). Recognition of oxPC CD36 also occurred regardless of whether the species were incorporated within the bilayer structure of a small unilamellar vesicle or embedded within a liposome derived from lipid extracts of native LDL, presumably at the hydrophobic/hydrophilic interface at the particle surface. Further, binding to oxPC CD36 -containing particles to CD36 increased with increasing mol % of ligand within the surface of a particle (Fig. 5c), consistent with enhancement of binding through mutivalent (i.e. multiple receptor-ligand) interactions. The MPO-H 2 O 2 -NO 2 Ϫ system was used as a reproducible mechanism for oxidizing lipids. The physiological relevance of this pathway for initiating lipid oxidation in vivo is supported by recent studies with leukocytes isolated from individuals with MPO deficiency (26) and by the recent demonstration that MPO serves as a strong and independent predictor of coronary artery disease risk (27).
It is interesting to note that small increases in oxPC-dependent vesicle uptake were also observed in CD36 null mice (Fig. 2). These results suggest that the incorporation of ox-PC CD36 species may also promote particle recognition by alternative scavenger receptors on MPM, but to a lesser extent than CD36. We speculate that the homologous scavenger receptor B1 is a likely candidate. It belongs to the same family of cell surface receptors as CD36, is expressed in macrophages (23), and recognizes modified lipoproteins similarly to CD36 (28). However, as our data demonstrate, the role of other receptors in recognizing oxPC CD36 is minor as compared with the role of CD36.
LC/ESI/MS/MS analyses of various oxPC species in fresh vascular tissues were performed using conditions developed to minimize intrapreparative oxidation of lipids and preserve recovery of each analyte monitored (based upon control studies using trace levels of individual synthetic standards and deuterated precursors, as described in Ref. 14). When analyzed using these methods, the content of oxPC CD36 species observed in vivo was comparable with those observed for the more terminally oxidized end products of PAPC, OV-PC, and G-PC (Fig.  4). The oxPC CD36 species identified are thus relatively abundant in vascular tissues. Moreover, like OV-PC and G-PC, which possess proinflammatory activities (19,29), oxPC CD36 species may possess biological activities extending beyond serving as ligands for CD36. In addition, by virtue of their capacity to carry cargo within oxLDL into CD36-expressing macrophages, oxPC CD36 may indirectly promote biological activities such as PPAR␥ activation (30 -32) or oxysterol-induced cytotoxicity (33) and apoptosis (34). Finally, it is tempting to speculate that individuals with increased circulating levels of ox-PC CD36 in plasma lipoproteins are at an increased risk for atherosclerotic lesion development and progression. The demonstration of structurally defined oxidized PC species that serve as apparent endogenous ligands for the macrophage scavenger receptor CD36 provides new opportunities for probing the physiological and pathophysiological activities of this complex and multifaceted receptor, as well as for assessing the role of oxidation in atherogenesis.