Determinants of Bioactivity of Oxidized Phospholipids
Specific Oxidized Fatty Acyl Groups at the sn-2 Position
Abstract—We previously described 3 bioactive oxidation products of 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine (PAPC) containing oxovaleroyl (POVPC), glutaroyl (PGPC), and epoxyisoprostane (PEIPC) groups at the sn-2 position that were increased in minimally modified/oxidized low density lipoprotein (MM-LDL) and rabbit atherosclerotic lesions. We demonstrated specific and contrasting effects of POVPC and PGPC on leukocyte-endothelial interactions and described an effect of PEIPC on monocyte binding. The major purpose of the present study was to determine the effects of structural changes on the bioactivities of these 3 lipids. We demonstrate herein that the group at the sn-2 position determines the specific bioactivity and that the substitution of stearoyl for palmitoyl at the sn-1 position or ethanolamine for choline at the sn-3 position of the phospholipid did not alter bioactivity. Oxidized PAPC, oxidized 1-stearoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine, and oxidized 1-stearoyl-2-arachidonoyl-sn-glycero-3-phosphorylethanolamine stimulated monocyte binding and inhibited lipopolysaccharide-induced expression of the neutrophil-binding molecule E-selectin. Furthermore, all oxovaleroyl phospholipids but not the glutaroyl phospholipids induced monocyte binding without an increase in vascular cell adhesion molecule-1 (VCAM-1) expression and inhibited lipopolysaccharide-induced E-selectin expression. In contrast, glutaroyl phospholipids but not oxovaleroyl phospholipids stimulated E-selectin and VCAM-1 expression. We further demonstrate that all parts of the phospholipid molecules are required for these bioactivities. Hydrolysis with phospholipase (PL) A1, PLA2, and PLC strongly reduced the bioactivities of POVPC, PGPC, and mixed isomers of PEIPC. PLD had a smaller but still significant effect. The effects of POVPC and PEIPC could be abolished by sodium borohydride treatment, indicating the importance of the reducible groups (carbonyl and epoxide) in these molecules. In summary, these studies identify 6 new bioactive, oxidized phospholipids that are increased in MM-LDL and, where measured, in atherosclerotic lesions. They thus suggest that a family of phospholipid oxidation products containing oxovaleroyl, glutaroyl, and epoxyisoprostane at the sn-2 position play an important role in the regulation of leukocyte-endothelial interactions, bioactivity being in part controlled by several types of phospholipid hydrolases.
- Received March 30, 2000.
- Accepted June 19, 2000.
Minimally oxidized LDL (MM-LDL), in contrast to native LDL, has been shown to stimulate endothelial cells to increase monocyte-endothelial interactions.1 Monocytes are the major inflammatory cells in the atherosclerotic lesion and have been shown to be important mediators of lesion initiation and progression.2 3 An important characteristic of atherosclerotic lesions and other chronic, inflammatory lesions is that they contain essentially no neutrophils.4 Our group has shown that the major bioactive lipids in MM-LDL were derived from oxidation of arachidonoyl phospholipids.5 6 We have identified 3 bioactive, oxidized phospholipids present in oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine (PAPC) as 1-palmitoyl-2-oxovaleroyl-sn-glycero-3-phosphorylcholine (POVPC), 1-palmitoyl-2-glutaroyl-sn-glycero-3-phosphorylcholine (PGPC), and 1-palmitoyl-2-(5,6-epoxyisoprostane E2)-sn-glycero-3-phosphorylcholine (PEIPC).7 8 The levels of these lipids were increased in atherosclerotic lesions in rabbit aortas, and natural antibodies to these oxidized phospholipids were present in the sera of apo E–null mice.8 9
We have observed several differences in action between POVPC and PGPC. POVPC induced monocyte binding and connecting segment-1 fibronectin expression, increased the levels of cAMP, and inhibited lipopolysaccharide (LPS)-induced E-selectin expression and neutrophil binding. The action of POVPC was inhibited by the platelet-activating factor receptor antagonist WEB 2086.10 11 In contrast, PGPC induced both monocyte and neutrophil binding and vascular cell adhesion molecule-1 (VCAM-1) and E-selectin expression, had no effect on connecting segment-1 or cAMP, and was not inhibited by WEB 2086.10 11 We further demonstrated that the effects of POVPC dominated in mixtures of the active compounds.
This report examines the molecular requirements for the specific actions of POVPC, PGPC, and PEIPC. We also identify new bioactive phospholipids with oxovaleroyl, glutaroyl, and epoxyisoprostane groups, demonstrating their role in the regulation of endothelial-leukocyte interactions. We determined that the levels of these lipids are increased in MM-LDL and atherosclerotic lesions.
Dimyristoyl-sn-glycero-3-phosphorylcholine (DMPC), dimyristoyl-sn-glycero-3-phosphorylethanolamine (DMPE), BHT, PMSF, soybean lipoxygenase, phospholipase (PL) A2 (from Naja naja venom), PLC (type IX, from Bacillus cereus), and PLD (type I, from cabbage) were obtained from Sigma Chemical Co. PLA1 (lipase from Rhizopus arrhizus) was obtained from Boehringer Mannheim. PAPC, 1-stearoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine (SAPC), and heart bovine PE fraction were obtained from Avanti Polar Lipids. Glutaric acid and NaBH4 were obtained from Aldrich Chemical Co.
Lipid and Lipoprotein Modification
LDL (1.019 to 1.069 g/mL) was isolated from the sera of normal blood donors by density gradient ultracentrifugation as described previously13 and stored at 4°C until use within 1 to 4 weeks of isolation. To prepare MM-LDL, LDL was pretreated with PMSF (3 mmol/L) and incubated at 1 milligram protein per milliliter in PBS at 37°C with Sepharose-bound soybean lipoxygenase (20 000 U) by using a previously described method.6 PMSF did not affect the levels of native and oxidized phospholipids measured in native LDL (data not shown). To isolate SAPE, bovine heart PE fraction was treated with 0.5N HCl for 15 minutes, extracted with chloroform/methanol (2:1, vol/vol), and then dried under argon.12 The residue was resuspended in chloroform and subjected to normal-phase coupled high-performance liquid chromatography–mass spectrometry (HPLC/MS, described below) to provide enriched 1-stearoyl-2-arachidonoyl-sn-glycero-3-phosphorylethanolamine (SAPE; charge-to-mass ratio [m/z] 768.5). PAPC, SAPC, and SAPE were auto-oxidized as described previously.8 POVPC, PGPC, SOVPC, and SGPC were prepared by ozonolysis of PAPC and SAPC by using the previously reported method.10 14 PEIPC was isolated from oxidized PAPC, and epoxyisoprostane (the oxidized fatty acid present in PEIPC) was generated by PLA2 hydrolysis of PEIPC as described previously.7 The concentration of bacterial endotoxin in each treatment solution was <20 pg/mL.
Analysis of Phospholipids
Normal-phase HPLC/MS analysis for oxidized PAPC and oxidized SAPC was performed as described previously,8 and oxidized SAPE was fractionated with acetonitrile/methanol/water (88:4:8, vol/vol/vol) containing 1 mmol/L ammonium acetate. PE derivatives were less stable than the corresponding PC derivatives and thus, were used for testing immediately after isolation and quantification. An API III triple-quadrupole biomolecular mass analyzer (Perkin-Elmer Sciex Instruments) was used for mass analysis and quantification of PC derivatives as described previously.8 For flow-injection analysis in the negative mode, phospholipids were dissolved in 100 μL of methanol/water (50:50, vol/vol) containing 10 mmol/L ammonium acetate and analyzed in the same solvent. For quantitative analysis, DMPC at 1 μg/10 mg tissue, 10 μg/mg lipoprotein was used as an internal standard for PC derivatives, and DMPE at 1 μg/mg lipoprotein was used as an internal standard for PE derivatives.15 To validate the use of DMPC as an internal standard for quantitative analyses, we correlated the levels of various phospholipids (PAPC, oxidized PAPC, POVPC, PGPC, and mixed isomers of PEIPC) determined by electrospray ionization–mass spectometry analysis, and their levels determined by phosphorous assay.16 The values obtained by the 2 methods differed by <5% for all chemical species except PEIPC, which differed by <10%, thus demonstrating reliable quantification of these lipids with DMPC. Similarly, we used DMPE as an internal standard for the analysis of PE derivatives.
PL Hydrolysis and NaBH4 Reduction of Oxidized Phospholipids
Oxidized phospholipids were hydrolyzed by various PLs by using a previously reported method with some modifications.7 17 To standardize the hydrolysis conditions, PAPC, oxidized PAPC, and PGPC were treated with different amounts of enzymes (1, 5, and 10 U/mL, with units defined by the supplier for the standard substrates) for different times (30 minutes and 1, 6, and 18 hours). On the basis of these studies, oxidized phospholipids were treated with 10 U/mL of PLs (PLA1, PLA2, PLC, or PLD) in PBS containing 5 mmol/L CaCl2 at pH 5.6, 7.0, 7.3, and 5.6, respectively). The samples were incubated at 37°C for 1 hour for PLA2 or overnight for PLA1, PLC, and PLD7 17 to achieve 70% to 100% hydrolysis. For NaBH4 reduction, oxidized phospholipids were treated with NaBH4 (50 mmol/L) in PBS or acetonitrile for 30 minutes7 8 to achieve 90% reduction. Phospholipid hydrolysis and NaBH4 reduction were monitored by the disappearance of starting phospholipids and/or the formation of hydrolysis or reduced products.
Cell Culture, Monocyte Adhesion Assay, and E-Selectin ELISA
Monocyte binding assays were performed as described previously1 with human aortic endothelial cells (HAECs) and human monocytes, except that the cells were treated overnight with 0.8 mg/mL lipoprotein-deficient serum and binding was performed in lipoprotein-deficient serum to reduce basal HAEC activation. The cells were then rinsed twice with medium, and binding of human monocytes to HAECs was performed essentially as described previously.1 VCAM-1 and E-selectin expression was measured on HAECs in 96-well dishes as described previously with the use of 10% fetal bovine serum during the incubation with oxidized phospholipids.10
Preparation of Lipids From Rabbit Aortas and Lipoproteins
The rabbits used in this study were a subset of animals used in another study reported elsewhere.18 Some rabbits (n=4) were studied while they consumed a standard chow diet, whereas other rabbits (n=6) were studied after 22 weeks on a semipurified, atherogenic diet with casein, sucrose, and butter as sources of protein, carbohydrate, and fat, respectively.18 At the end of treatment with the atherogenic diet, the rabbits were exsanguinated after ketamine hydrochloride anesthesia (6 mg/kg body weight). Aortas were removed, cleaned of adventitial tissue (at 4°C), weighed, frozen at −70°C under argon, and stored protected from light.18 Total lipids and phospholipids were isolated from LDL, MM-LDL, and rabbit aortas as described previously.8
Data were analyzed by 1-way ANOVA, probability values were calculated by using StatView (Abacus Concepts, Inc), and P<0.05 was considered statistically significant.
Comparison of Oxidation Products in Oxidized PAPC, Oxidized SAPC, and Oxidized SAPE
We first tested the hypothesis that oxidation of SAPC and SAPE would produce oxidation products similar to those in oxidized PAPC. Auto-oxidation of PAPC generated a complex mixture of products, including 3 bioactive, oxidized phospholipids, POVPC (m/z 594.3), PGPC (m/z 610.2), and mixed isomers of PEIPC (m/z 828.5).7 8 Here, we report that auto-oxidation of SAPC and SAPE also resulted in a similar addition of molecular oxygen or fragmentation of the arachidonate moiety at the sn-2 position of the phospholipid (online Figures I through III; please see www… ). These results suggest that the changes at the sn-1 and sn-3 groups did not affect the oxidation patterns of arachidonoyl phospholipids. Although the m/z values of the derivatives in oxidized SAPC and oxidized SAPE suggested that they represented oxovaleroyl, glutaroyl, and epoxyisoprostane phospholipids, we confirmed this identification by using tandem mass spectrometry. Tandem mass spectrometry analyses of phospholipids generated characteristic fragment ions corresponding to the sn-1, sn-2 fatty acid groups, and an ion of lysophospholipid.8 For example, the ion corresponding to 1-stearoyl-2-oxovaleroyl-sn-glycero-3-phosphorylcholine (SOVPC, MH+ ion with m/z 622.3) in oxidized SAPC (online Figure⇓ II; please see www… ) produced an (M-CH3)− ion at m/z 606.3 in the negative mode, which gave characteristic fragments at m/z 115 (oxovalerate), 283.1 (stearate), and 508.2 [(M-CH3)− ion of lysophospholipid]. This result confirmed that the ion at m/z 622.3 in oxidized SAPC was SOVPC. Similar analyses for ions corresponding to bioactive, oxidized phospholipids in oxidized SAPC and oxidized SAPE gave the characteristic sn-2 fragments at m/z 115 (oxovalerate), 131 (glutarate), or 349 (epoxyisoprostane) along with ions for sn-1 fatty acids and lysophospholipids. These analyses confirmed the structure of molecules in oxidized SAPC as SOVPC (m/z 622.3), SGPC (m/z 638.2), and mixed isomers of SEIPC (m/z 856.6) and in oxidized SAPE as SOVPE (m/z 580.2), SGPE (m/z 596.2), and mixed isomers of SEIPE (m/z 814.6).
Effect of Oxidized Phospholipids on Monocyte Binding
Treatment of endothelial cells with oxidized arachidonoyl phospholipids (PAPC, SAPC, and SAPE at 5 to 50 μg/mL) for 4 hours at 37°C induced binding of monocytes in a dose-dependent fashion. The corresponding native arachidonoyl phospholipids (50 μg/mL) were inactive (Figure 1⇑). We fractionated oxidized SAPC and oxidized SAPE by normal-phase HPLC/MS and tested the fractions for their ability to stimulate endothelial cells to bind monocytes. The fractions containing the molecules with m/z values corresponding to oxovaleroyl, glutaroyl, and epoxyisoprostane derivatives (determined by reconstructed selective-ion chromatograms) stimulated endothelial cells to bind monocytes. The bioactivity of oxidized phospholipids present in oxidized SAPC and oxidized SAPE was further confirmed by dose-response studies with known amounts of isolated and synthetic molecules. SOVPC, SGPC, and mixed isomers of SEIPC (m/z 856.6) were isolated from oxidized SAPC, and fractions that were enriched in SOVPE, SGPE, and mixed isomers of SEIPE (m/z 814.5) were isolated from oxidized SAPE and tested for their ability to induce monocyte binding. These isolated molecules showed a dose-dependent increase in monocyte binding. Synthetic SOVPC and SGPC also showed a similar dose-dependent increase in monocyte binding similar to the corresponding isolated molecules (Figure 2⇓).
Effect of Oxidized Phospholipids on E-Selectin and VCAM-1 Expression
As was previously reported for oxidized PAPC, oxidized SAPC, and oxidized SAPE, all (5 to 50 μg/mL) effectively inhibited the action of LPS on E-selectin expression in a dose-dependent fashion, whereas the corresponding native arachidonoyl phospholipids (50 μg/mL) did not have any significant effect (Figure 3⇓). Oxovaleroyl phospholipids alone caused a similar inhibition. POVPC, SOVPC, and SOVPE (1 to 5 μg/mL) dose-dependently inhibited LPS-induced E-selectin expression (online Figure⇑ IV; please see www… ). In contrast, PGPC was previously shown to induce E-selectin, VCAM-1 expression, and binding of polymorphonuclear neutrophils to HAECs.10 Here, we demonstrate that PGPC, SGPC, and SGPE all induced E-selectin and VCAM-1 on HAECs (Figure 4⇓). However, oxovaleroyl phospholipids alone did not induce significant levels of VCAM-1 and E-selectin on HAECs (data not shown). These results showed that the changes at the sn-1 and sn-3 positions did not affect the ability of oxovaleroyl phospholipids to inhibit LPS action and glutaroyl phospholipids to stimulate E-selectin or VCAM-1 expression.
Comparison of Phospholipids in LDL and MM-LDL
POVPC, SOVPC, PGPC, SGPC, mixed isomers of PEIPC, and mixed isomers of SEIPC were increased by 2- to 11-fold in MM-LDL phospholipids compared with native LDL phospholipids (the Table⇓ , columns 1 and 2). The levels of SOVPE, SGPE, and mixed isomers of SEIPE were increased by 3- to 5-fold in MM-LDL lipids compared with native LDL lipids. The levels of bioactive PC derivatives detected were ≈10- to 25-fold higher than those of the corresponding PE derivatives. The bioactive, oxidized phospholipids represent a small fraction of the arachidonoyl phospholipid oxidation products (5% to 10%). However, this amount is sufficient to account for the bioactivity of MM-LDL.
Comparison of Phospholipid Levels in Normal Rabbit Aorta and Rabbit Atherosclerotic Lesions
Cholesterol, PAPC, and SAPC were all increased in the rabbits fed the atherogenic diet. The levels of bioactive, oxidized phospholipids, POVPC, PGPC, mixed isomers of PEIPC, SOVPC, SGPC, and mixed isomers of SEIPC were increased by 3- to 6-fold in animals fed the atherogenic diet (the Table⇑ , columns 3 and 4). These levels correspond to ≈116, 62, and 85 μg/mL POVPC, PGPC, and mixed isomers of PEIPC, respectively (by approximate conversion of wet weight to volume). This level is roughly 10 to 20 times that required to activate the endothelium.
Effects of PLs and NaBH4 on the Bioactivity of Oxidized Phospholipids
To further identify the structural motifs of oxidized phospholipids that are important for their bioactivity, we tested the effects of chemical modification and enzymatic digestion. Treatment of oxidized phospholipids with NaBH4 reduced the oxygenated groups such as aldehydes, ketones, and epoxides.7 8 The specific PLs PLA1, PLA2, PLC, and PLD were shown to selectively hydrolyze the sn-1, sn-2, and sn-3 groups and the phosphate-choline bond, respectively19 (online Figure⇑ V; please see www… ). In the present study, we tested the effects of PL and NaBH4 treatments on the bioactivity of POVPC, PGPC, and mixed isomers of PEIPC by a monocyte binding assay (Figure 5⇓). Treatment of these 3 active lipids with PLA1, PLA2, and PLC completely abolished their bioactivity, whereas PLD partially reduced their bioactivity. NaBH4 treatment abolished the activity of POVPC and mixed isomers of PEIPC; however, the activity of PGPC was unaltered (Figure 5⇓). These studies suggest that the aldehyde group in POVPC and reducible groups in PEIPC (carbonyl and epoxide) are important for their bioactivity. Taken together, these studies show that the sn-1, sn-2, and sn-3 groups are essential for their effects on monocyte binding; however, hydrolysis with PLD only partially reduced activity. The ability of POVPC (5 μg/mL) to inhibit LPS-induced E-selectin expression was strongly reduced by all 4 PLs and NaBH4 (online Figure⇑ VI; please see www… ). Similarly, PGPC- (5 μg/mL) induced E-selectin expression was also significantly reduced by all 4 PLs; however, NaBH4 had no significant effect (online Figure⇑ VI; please see www… ).
The common structural motifs for the bioactive, oxidized phospholipids are the presence of oxovaleroyl, glutaroyl, and an epoxyisoprostane at the sn-2 position; hence, we examined the effects of the isolated sn-2 fatty acids on monocyte adhesion. Oxovaleric acid (1 to 10 μmol/L, fatty acid from POVPC), glutaric acid (1 to 10 μmol/L, fatty acid from PGPC), and epoxyisoprostane (1 to 3 μmol/L, fatty acid from PEIPC) did not induce monocyte binding at the concentrations at which the corresponding intact, oxidized, phospholipids are maximally active (data not shown).
These studies were designed to (1) examine the effects of structural changes and hydrolytic enzymes on the actions of bioactive, oxidized phospholipids and (2) to quantify the levels of these lipids in MM-LDL and in atherosclerotic lesions. In previous studies, we identified 3 oxidized phospholipids primarily responsible for the activity of oxidized PAPC.7 8 Although PAPC is the major arachidonoyl phospholipid in native human LDL, SAPC and SAPE are also present. We now provide evidence for the formation of bioactive, oxidized phospholipids from the oxidation of SAPC and SAPE and for their increased levels in MM-LDL (Figure 1⇑ and the Table⇑).
In the present study, we have presented evidence that the sn-2 position of the bioactive, oxidized phospholipids is the main determinant of the type of activity shown by the lipid. The changes in saturated fatty acid at the sn-1 position and in the head group at the sn-3 position did not affect bioactivity (Figures 1⇑ and 2⇑). We have demonstrated that oxovaleroyl-, glutaroyl-, and epoxyisoprostane-containing phospholipids all stimulated endothelial cells to bind monocytes (Figure 2⇑). Oxovaleroyl phospholipids were effective in inhibiting LPS induced E-selectin expression (online Figure⇑ IV; please see www… ). In contrast, glutaroyl phospholipids induced the surface expression of E-selectin and VCAM-1 in HAECs (Figure 4⇑). Thus, 2 types of oxidized phospholipids, whether derived from PAPC, SAPC, or SAPE, have quite different effects, and their relative levels could control the recruitment of specific leukocytes at the sites of inflammation.
The present study demonstrates that all 3 groups, sn-1, sn-2 and sn-3, of bioactive, oxidized phospholipids are important for bioactivities (Figure 5⇑). Our group has previously shown that the 2 hydrolytic enzymes, platelet-activating factor–acetyl hydrolase and paraoxonase, were able to destroy the bioactivity of oxidized phospholipids.5 6 Studies with NaBH4 demonstrate that the reducible functional groups (carbonyl and epoxide) are essential for bioactivity. However, certain changes at the sn-3 position (PC to PE) do not alter their activity, and cleavage of the choline moiety of PC by PLD partially reduced its bioactivity (Figures 4⇑ and 5⇑). The past and present studies show that neither lysophospholipids6 nor the sn-2 fatty acids in the free form were active in inducing monocyte binding. Lysophosphatidic acid has been shown to exert a variety of biological responses, including stimulation of endothelial cells to bind monocytes and neutrophils and to induce E-selectin and VCAM-1 expression.20 The formation of lysophosphatidic acid and/or lysophosphatidic acid analogues after treatment of oxidized phospholipids with PLD may explain the partial, and not complete, loss of bioactivity (Figure 5⇑).
Our present and past data demonstrate that the bioactive, oxidized phospholipids responsible for stimulating monocyte binding and inhibiting neutrophil binding are increased in atherosclerotic lesions. The presence of oxidized LDL in rabbit and human atherosclerotic lesions21 and increased levels of total phospholipids and lysophosphatidylcholine in rabbit lesions22 have been previously demonstrated. In the present study, the levels of oxidized phospholipids in individual aortas were highly correlated with the accumulation of cholesterol and unoxidized phospholipids (data not shown). One interpretation of this finding is that increased levels of phospholipid oxidation products in the vessel wall are due to the increase in phospholipid, which serves as the oxidative substrate. However, a number of other mechanisms may explain the increased oxidation observed. In general, our results suggest that these oxidized phospholipids in the vessel wall could serve as specific markers of oxidative injury in vivo.
In summary, these results identify 6 new bioactive, oxidized phospholipids that have specific effects on leukocyte–endothelial cell interactions. We have demonstrated that the specificity is determined by the sn-2 groups of oxidized phospholipids. The levels of all 6 bioactive, oxidized phospholipids are increased in MM-LDL, and the levels of at least 3 bioactive, oxidized phospholipids are increased in atherosclerotic lesions, suggesting a role for these newly identified bioactive, oxidized phospholipids in atherogenesis. Finally, these results demonstrate the importance of the integrity of the sn-1, sn-2, and sn-3 positions in the activation of endothelial cells.
This work was supported by US Public Health Service grant No. HL 30568, the W.M. Keck Foundation (to K.F.F.), and the Laubish Fund. We would like to thank Dr Mohamad Navab for helpful suggestions.
Berliner JA, Territo MC, Sevanian A, Ramin S, Kim JA, Bamshad B, Esterson M, Fogelman AM. Minimally modified low density lipoprotein stimulates monocyte endothelial interactions. J Clin Invest. 1990;85:1260–1266.
Berliner JA, Gerrity RG. Pathology of atherogenesis. In: Lusis AJ, Rotter JI, Sparkes RS, eds. Molecular Genetics of Coronary Artery Disease. Basel, Switzerland: S Karger; 1992:1–15.
Watson AD, Berliner JA, Hama SY, La Du BN, Faull KF, Fogelman AM, Navab M. Protective effect of high density lipoprotein associated paraoxonase: inhibition of the biological activity of minimally oxidized low density lipoprotein. J Clin Invest. 1995;96:2882–2891.
Watson AD, Navab M, Hama SY, Sevanian A, Prescott SM, Stafforini DM, McIntyre TM, Du BN, Fogelman AM, Berliner JA. Effect of platelet activating factor-acetylhydrolase on the formation and action of minimally oxidized low density lipoprotein. J Clin Invest. 1995;95:774–782.
Watson AD, Subbanagounder G, Welsbie DS, Faull KF, Navab M, Jung ME, Fogelman AM, Berliner JA. Structural identification of a novel pro-inflammatory epoxyisoprostane phospholipid in mildly oxidized low density lipoprotein. J Biol Chem. 1999;274:24787–24798.
Watson AD, Leitinger N, Navab M, Faull KF, Hörkkö S, Witztum JL, Palinski W, Schwenke D, Salomon RG, Sha W, Subbanagounder G, Fogelman AM, Berliner JA. Structural identification by mass spectrometry of oxidized phospholipids in minimally oxidized low density lipoprotein that induce monocyte/endothelial interactions and evidence for their presence in vivo. J Biol Chem. 1997;272:13597–13607.
Hörkkö S, Miller E, Dudl E, Reaven P, Curtiss LK, Zvaifler NJ, Terkeltaub R, Pierangeli SS, Branch DW, Palinski W, Witztum JL. Antiphospholipid antibodies are directed against epitopes of oxidized phospholipids: recognition of cardiolipin by monoclonal antibodies to epitopes of oxidized low density lipoprotein. J Clin Invest. 1996;98:815–825.
Leitinger N, Tyner TR, Oslund L, Rizza C, Subbanagounder G, Lee H, Shih PT, Mackman N, Tigyi G, Territo MC, Berliner JA, Vora DK. Structurally similar oxidized phospholipids differentially regulate endothelial binding of monocytes and neutrophils. Proc Natl Acad Sci U S A. 1999;96:12010–12015.
Subbanagounder G, Leitinger N, Shih PT, Faull KF, Berliner JA. Evidence that phospholipid oxidation products and/or platelet-activating factor play an important role in early atherogenesis: in vitro and in vivo inhibition by WEB 2086. Circ Res. 1999;85:311–318.
Havel R, Eder H, Bragdon J. The distribution and chemical composition of ultracentrifugally separated lipoproteins of human serum. J Biol Chem.. 1955;43:1345–1353.
Marathe GK, Davies SS, Harrison KA, Silva AR, Murphy RC, Castro-Faria-Neto H, Prescott SM, Zimmerman GA, McIntyre TM. Inflammatory platelet-activating factor-like phospholipids in oxidized low density lipoproteins are fragmented alkyl phosphatidylcholines. J Biol Chem. 1999;274:28395–28404.
Schwenke DC, Behr SR. Vitamin E combined with selenium inhibits atherosclerosis in hypercholesterolemic rabbits independently of effects on plasma cholesterol concentrations. Circ Res. 1998;83:366–377.
Prescott SM. Minireview prologue: a thematic series on phospholipases. J Biol Chem. 1997;272:15043.
Ylä-Herttuala S, Palinski W, Rosenfeld ME, Parthasarathy S, Carew TE, Butler S, Witztum JL, Steinberg D. Evidence for the presence of oxidatively modified low density lipoprotein in atherosclerotic lesions of rabbit and man. J Clin Invest. 1989;84:1086–1095.