Platelet Activation by Oxidized Low Density Lipoprotein Is Mediated by Cd36 and Scavenger Receptor-A

【摘要】  Objective— The interaction of platelets with low density lipoprotein (LDL) contributes to the development of cardiovascular disease. Platelets are activated by native LDL (nLDL) through apoE Receptor 2' (apoER2')-mediated signaling to p38 MAPK and by oxidized LDL (oxLDL) through lysophosphatidic acid (LPA) signaling to Rho A and Ca 2+. Here we report a new mechanism for platelet activation by oxLDL.

Methods and Results— Oxidation of nLDL increases p38 MAPK activation through a mechanism that is (1) independent of LPA, and (2) unlike nLDL-signaling not desensitized by prolonged platelet-LDL contact or inhibited by receptor-associated protein or chondroitinase ABC. Antibodies against scavenger receptors CD36 and SR-A alone fail to block p38 MAPK activation by oxLDL but combined blockade inhibits p38 MAPK 40% and platelet 60%. Mouse platelets deficient in either CD36 or SR-A show normal p38 MAPK activation by oxLDL but combined deficiency of CD36 and SR-A disrupts oxLDL-induced activation of p38 MAPK 70%.

Conclusion— These findings reveal a novel platelet-activating pathway stimulated by oxLDL that is initiated by the combined action of CD36 and SR-A.

Platelets are activated by nLDL through binding to apoER2', which signals to p38 MAPK. Oxidation induces more p38 MAPK activation through loss of apoER2' binding thereby initiating the combined activity of CD36 and SR-A.

【关键词】  platelets LDL oxidized LDL CD scavenger receptorA

Introduction

An elevated level of native low density lipoprotein (nLDL) is a risk factor for arterial thrombosis and atherosclerosis as demonstrated in familial hypercholesterolemia, where defective apoB/E receptors fail to remove nLDL from the circulation. Atherogenesis starts when nLDL accumulates in the vessel wall at sites of injury and is oxidized by products from macrophages, smooth muscle cells, and endothelial cells. 1 Oxidized LDL (oxLDL) accumulates in monocytes that have infiltrated the subendothelium and differentiated into macrophages. The resulting foam cells are characteristic for the early atherosclerotic lesion. 2 OxLDL further contributes to atherosclerosis because it contains lysophosphatidic acid (LPA), which starts platelet shape change and aggregation. 3

In healthy individuals, the concentration of oxLDL is low. The normal intima contains little oxLDL (1.86±0.59 ng/µg apolipoprotein B100 [apoB100]) but levels increase 6-fold in atherosclerotic lesions (11.9±1.7 ng/µg apoB100). 4 Blood from atherosclerotic patients contains autoantibodies that react with oxidation-specific epitopes in both the lipid and protein moiety of oxLDL, 5,6 indicating that oxLDL is also present in the circulation. Hence, in the circulation, platelets can come into contact with oxLDL and become activated, thereby contributing to thrombotic occlusion.

The oxidation of nLDL in vivo can be mimicked in vitro by treatment of nLDL with FeSO 4. These oxLDL preparations resemble in vivo oxidized LDL with respect to electrophoretic mobility, density, LPA content, fragmentation of apoB100, chemotactic activity for monocytes, and susceptibility to degradation by macrophages. 3,7–9 LPA makes oxLDL a potent platelet activating agent 3 through activation of its LPA 1 and LPA 3 receptors, 10 which are members of the endothelial differentiation gene receptor family. At low concentrations, LPA stimulates Rho, Rho-kinase, and myosin light chain phosphorylation resulting in platelet shape change caused by changes in the actin cytoskeleton. 11 At high concentrations, LPA stimulates Ca 2+ mobilization and the tyrosine kinases Syk and Src resulting in integrin IIb β 3 activation and aggregation. 12

Recently, our laboratory identified the signaling receptor through which nLDL changes the behavior of platelets: the apolipoprotein E Receptor 2' (apoER2'). 13 It is a 130-kDa splice variant of apoER2, a member of the LDL receptor family, also known as LRP8. The ligand-binding domain of full-length apoER2 contains 7 complement type A binding repeats 14 and the apoER2' variant lacks binding repeats 4 to 6. 15 ApoER2' is activated by contact with the receptor-specific domain within apoB100 of LDL, called the B-site, with amino acid sequence RLTRKRGLKLA. 13 Receptor activation starts signaling through p38 MAPK and cytosolic phospholipase A 2 leading to formation of thromboxane A 2, a potent platelet activating agent. 16 The result is an increase in responsiveness to thrombin, collagen, and ADP, leading to enhanced aggregation and secretion on contact with nLDL.

To understand how oxidation enhances the platelet-activating properties of nLDL, signaling through p38 MAPK has been compared with LPA-dependent platelet activation at different degrees of LDL oxidation. 17 Both pathways were mutually independent because oxLDL signaling to p38 MAPK was unaffected by L-NASPA, an inhibitor of LPA receptors, 17 and LPA was incapable of activating p38 MAPK. 16 Below 30% oxidation, activation occurred primarily through the p38 MAPK pathway resulting in sensitization, whereas at higher oxidation levels also the LPA pathway was initiated resulting in aggregation. 17

In an attempt to understand platelet activation by oxLDL in more detail, we compared the activation of the p38 MAPK pathway by oxLDL with that induced by nLDL. We found that oxidation led to a strong increase in p38 MAPK signaling caused by loss of apoER2' activation and appearance of the combined activity of CD36 and scavenger receptor-A.

Methods

To elucidate the receptors mediating platelet activation by oxLDL, human and murine platelets from wild-type C57Bl/6 mice and mice deficient in either CD36, SR-A, or both were incubated with 1.0 g/L native or partially oxidized LDL and phosphorylation of apoER2' and/or p38 MAPK was determined as a measure of activation. In these experiments, nLDL and oxLDL signaling to p38 MAPK was separated by prolonged incubation of platelets with nLDL before treatment with nLDL or partially oxidized LDL. To investigate functional consequences of oxLDL signaling through CD36 and SR-A, human platelets were incubated with partially oxidized LDL (31% to 60% oxidation) and adhesion to immobilized fibrinogen was measured at a shear rate of 300 s –1 in the absence and presence of inhibitors.

A full description of the preparation of native and modified LDL, the isolation of human and murine platelets, measurement of phosphorylation of apoER2' and p38 MAPK, and adhesion to immobilized fibrinogen is available in the detailed Methods section at http://atvb.ahajournals.org.

Results

P38 MAPK Signaling by oxLDL and nLDL

To understand how oxidation changes the regulation of the p38 MAPK pathway, LDL was oxidized for 30% to 60% and compared with nLDL. At saturating concentrations (1.0 g/L), partially oxidized LDL induced about 2-fold more p38 MAPK phosphorylation than nLDL without affecting the transient kinetics of this activation. ( Figure 1A and 1 B). This agrees with earlier observations showing 60% led to a proportional increase in p38 MAPK activation. 17 Thus, oxidation increases activation of the p38 MAPK pathway. Recently, we described that nLDL activates the p38 MAPK pathway through apoER2' thereby increasing the sensitivity of platelets to agonist stimulation. 13 To determine whether the enhanced activation of p38 MAPK by partially oxidized LDL was the result of better activation of apoER2', platelets were incubated with both types of LDL and tyrosine phosphorylation of apoER2' was measured. Both LDL preparations induced a similar activation of apoER2' reaching a maximum after 30 seconds and returning to prestimulation values after 10 minutes ( Figure 1 C). Thus, the extra activation of p38 MAPK by partially oxidized LDL could not be attributed to stronger apoER2' activation.

Figure 1. P38 MAPK signaling by oxLDL and nLDL. Platelets were incubated with nLDL or partially oxidized LDL (31% to 60% oxidation) at 37°C for the indicated time periods (1.0 g/L (A)) or at the indicated concentrations (1 minute) (B) and p38 MAPK phosphorylation was determined. Data were expressed as percentage of the density of incubation with nLDL at 1 minute. C, Platelets were incubated with 1.0 g/L nLDL or partially oxidized LDL. At the indicated time points, apoER2' was immunoprecipitated from platelet lysates and tyrosine phosphorylation was detected. Data were expressed as percentage of the density after 30 seconds of incubation with nLDL. Means±SEM, n=3.

Oxidation of nLDL Introduces ApoER2'-Independent Platelet Signaling

To identify the pathway through which oxLDL activates p38 MAPK, platelets were first treated with 1.0 g/L nLDL for 1 minute to saturate apoER2' signaling to p38 MAPK and thereafter treated with a second dose of either nLDL or partially oxidized LDL. A second addition of nLDL did not further increase p38 MAPK phosphorylation, confirming maximal activation by the first dose. In contrast, a second addition of a saturating concentration of partially oxidized LDL increased p38 MAPK activity 4-fold to the range found with a single dose of oxLDL (supplemental Figure IA). A second addition of nLDL or oxLDL after a first treatment with oxLDL did not change p38 MAPK activation by oxLDL. These data fit to the concept that oxidation of LDL introduces a second property that activates p38 MAPK and is independent of the activation induced by nLDL.

Receptor-associated protein (RAP) is a specific blocker of ligand binding to LDL receptor family members. RAP induced a dose-dependent inhibition of p38 MAPK phosphorylation induced by nLDL, which is in agreement with involvement of apoER2', an LDL receptor family member (supplemental Figure IB). In contrast, p38 MAPK phosphorylation induced by oxLDL was hardly affected by RAP, the minor decrease probably reflecting residual nLDL in this partially oxidized preparation. This conclusion was supported by the observation that apoER2' tyrosine phosphorylation induced by nLDL and oxLDL was equally inhibited by RAP (supplemental Figure IC). Chondroitinase ABC removes chondroitin and dermatan sulfate side chains from proteoglycans, which are important for nLDL binding to its receptor. 18 Chondroitinase ABC completely abolished p38 MAPK phosphorylation by nLDL, but hardly interfered with oxLDL-induced phosphorylation, the minor decrease again probably resulting from residual nLDL (supplemental Figure ID). Treatment with L-NASPA to block LPA binding to its receptors did not affect p38 MAPK activation by oxLDL (data not shown). Thus, the receptor that mediates the extra p38 MAPK activation by oxLDL is therefore not a member of the LDL receptor or LPA receptor families.

Desensitization of nLDL-Induced Signaling

A property of many receptors is the ability to become desensitized after prolonged ligand contact. To assess whether receptors mediating nLDL and oxLDL signaling to p38 MAPK could be desensitized, platelets were incubated with nLDL (1.0 g/L) for 30 minutes and thereafter stimulated with a second dose of nLDL or with partially oxidized LDL. Indeed, a first treatment with nLDL almost completely abolished p38 MAPK activation by a second dose of nLDL, indicating that this treatment arrested nLDL signaling ( Figure 2 A). In contrast, p38 MAPK activation by partially oxidized LDL remained mostly intact after a first treatment with nLDL, the minor decrease reflecting residual nLDL ( Figure 2 A). Figure 2 B illustrates the desensitization of nLDL signaling by preincubation with nLDL and residual nLDL present in 2 partially oxidized LDL preparations. OxLDL signaling was not desensitized. Thus, prolonged contact with nLDL desensitized nLDL signaling but not oxLDL signaling to p38 MAPK.

Figure 2. Desensitization of nLDL-induced signaling. A, Platelets were treated with vehicle (left panel) or nLDL (1.0 g/L, 30 minutes, right panel), stimulated with 1.0 g/L nLDL or partially oxidized LDL and p38 MAPK 60% oxidation) for 30 minutes, stimulated with vehicle, 1.0 g/L nLDL, or partially oxidized LDL and p38 MAPK phosphorylation was determined. C, Platelets were treated with vehicle or 1.0 g/L nLDL for 30 minutes, then stimulated with 1.0 g/L LDL oxidized to the indicated extents (1 minute) and phosphorylation of p38 MAPK was determined. D, Platelets were incubated with vehicle (left panel) or nLDL (1.0 g/L, 30 minutes, right panel), again stimulated with 1.0 g/L nLDL (1 minute), and tyrosine phosphorylation of apoER2' was measured. A and D, Data were expressed as percentage of the density of incubations with 1 minute nLDL before desensitization. C, Data were expressed as percentage of the density of incubations with oxLDL before desensitization (white bars=100%). Means±SEM, n=3.

The possibility to separate nLDL and oxLDL signaling to p38 MAPK by prolonged incubation with nLDL was used to quantify the contribution of oxidation to the signaling properties of LDL. nLDL preparations were oxidized to different extents, and p38 MAPK was activated before and after the desensitization phase induced by nLDL ( Figure 2 C). nLDL contained <15% oxidized LDL and was responsible for 90% of total p38 MAPK phosphorylation through the nLDL pathway. As more LDL was oxidized, the contribution of nLDL signaling to p38 MAPK decreased and was replaced by oxLDL signaling. LDL 60% oxidation activated the platelets almost exclusively through the oxLDL receptor.

The loss of nLDL signaling after prolonged nLDL-platelet contact appears a logical consequence of desensitization at the level of the receptor. To confirm this concept, apoER2' phosphorylation by nLDL was analyzed before and after 30 minutes incubation with nLDL ( Figure 2 D). Unexpectedly, a similar receptor activation was observed, illustrating that ligand activation of apoER2' is fully reversible. Apparently, the cause of desensitization of p38 MAPK signaling is not at the level of the receptor and must be sought downstream of apoER2'.

CD36 and Scavenger Receptor-A Mediate oxLDL-Induced Activation of p38 MAPK in Human and Murine Platelets

Scavenger receptors (SRs) are membrane glycoproteins that bind oxLDL on cells like macrophages, smooth muscle cells, and platelets. 19 To investigate whether SRs play a role in p38 MAPK activation, platelets were incubated with antibody FA6.152 against CD36, 20 which is a class-B SR and also known as glycoprotein IV, and with fucoidan, which inhibits oxLDL binding to class A SRs. 21 Subsequent stimulation with nLDL as a control and with partially oxidized LDL (31% to 60% oxidation) showed that each treatment failed to change the p38 MAPK activation by both nLDL and oxLDL. In contrast, combined incubation with both inhibitors sharply decreased p38 MAPK 40% to the level found with nLDL ( Figure 3 A). Thus, the combined blockade of CD36 and SR-A leads to complete inhibition of oxLDL signaling to p38 MAPK leaving nLDL signaling to p38 MAPK undisturbed. The specificity of this inhibition was confirmed by the inability of RAP to interfere with oxLDL-induced signaling and of fucoidan with nLDL-induced signaling ( Figure 3B and 3 C).

Figure 3. CD36 and SR-A mediate oxLDL-induced activation of p38 MAPK in human platelets. A, Platelets were treated with vehicle, FA6.152 (anti-CD36; 4 µg/mL, 30 minutes), and fucoidan (anti–SR-A; 50 µg/mL, 30 minutes), stimulated with nLDL or partially oxidized LDL (1.0 g/L, 1 minute), and p38 MAPK phosphorylation was determined. Data were expressed as percentage of the density of incubations with nLDL. Means±SEM, n=3. B and C, Platelets were treated with vehicle, (B) GST-RAP (50 µg/mL, 10 minutes), or (C) fucoidan, stimulated with the indicated concentrations (B) partially oxidized LDL or (C) nLDL (1 minute), and p38 MAPK phosphorylation was determined.

To confirm these observations by targeted interference with receptor expression, incubations with oxLDL were repeated with murine platelets deficient in either CD36 (CD36 –/– ), SR-A (SR-A –/– ), or both (CD36 x SR-A double knockout ). Murine wild-type (w.t.) platelets showed the same p38 MAPK activation by nLDL and oxLDL as human platelets and again only the combined addition of FA6.152 and fucoidan brought oxLDL 70% back to levels found with nLDL ( Figure 4 ). A similar blockade of oxLDL signaling was induced in CD36 –/– platelets incubated with fucoidan and in SR-A –/– platelets incubated with FA6.152. The combined deletion of CD36 and SR-A expression sharply reduced oxLDL signaling even in the absence of additions. These results support the findings in human platelets and show that interference with both CD36 and SR-A is required for full inhibition of oxLDL signaling to p38 MAPK.

Figure 4. CD36 and SR-A mediate oxLDL-induced activation of p38 MAPK in murine platelets. WT murine platelets, CD36 –/–, SR-A –/–, and CD36 x SR-A dKO platelets were treated with vehicle, FA6.152, or fucoidan before stimulation with 1.0 g/L nLDL or partially oxidized LDL, and p38 MAPK phosphorylation was determined. Data were expressed as percentage of the density of incubations with nLDL Means±SEM, n=3. * P <0.05, ** P <0.01.

CD36 and SR-A Mediate oxLDL-Induced Platelet Adhesion to Immobilized Fibrinogen

To investigate functional consequences of oxLDL signaling through CD36 and SR-A, human platelets were incubated with oxLDL (31% to 60% oxidation) and adhesion to immobilized fibrinogen was measured at a shear rate of 300 s –1. OxLDL increased adhesion with about 35%, probably as a result of activation through the p38 MAPK pathway because the p38 MAPK inhibitor SB203580 strongly reduced adhesion ( Figure 5 B). At a higher shear rate (1200 s –1 ), resembling the shear rates encountered in the arterial circulation, oxLDL induced a similar increase (24%) in platelet adhesion to fibrinogen (data not shown). Although the anti-CD36 antibody FA6.152 alone had no effect on oxLDL-induced p38 MAPK activation, it inhibited platelet adhesion to fibrinogen by 37% to the level found in the absence of oxLDL. The presence of fucoidan alone had no effect, but in combination with FA6.152 there was a reduction of 63% ( Figure 5 B). Thus, adhesion to fibrinogen under flow is enhanced by oxLDL signaling in which both CD36 and SR-A participate and which appears to be a direct result of increased p38 MAPK activation through these receptors.

Figure 5. CD36 and SR-A mediate oxLDL-induced platelet adhesion to immobilized fibrinogen. Reconstituted blood containing platelets was treated with vehicle or (A, B) FA6.152, fucoidan, SB203580 (anti-p38 MAPK; 10 µmol/L, 30 minutes, 37°C), or (C) AR-C69931MX (anti-P2Y 12 receptor, 1 µmol/L, 5 minutes, 37°C) before stimulation with partially oxidized LDL (0.2 g/L, 5 minutes, 37°C). The blood was perfused over immobilized fibrinogen for 5 minutes at a shear rate of 300 s –1. Original magnification x 400. B and C, Platelet adhesion was expressed as the percentage of the surface covered with platelets compared with nontreated platelets (100%). Means±SEM, n=3. * P <0.05, ** P <0.01.

To investigate the role of feedback activation by secreted ADP, the experiment was repeated in the presence of the P2Y 12 receptor antagonist AR-C69931MX. This treatment reduced adhesion from 100% to 50%, confirming earlier observations. 22 In the presence of oxLDL, platelet adhesion was reduced from 135% to 75%. Thus, also in platelets with blocked P2Y 12 signaling, the stimulation of adhesion by oxLDL was preserved ( Figure 5 C).

Discussion

The present study shows that oxLDL activates p38 MAPK through the combined stimulation of CD36 and SR-A thereby increasing platelet adhesion to fibrinogen under flow. This property functions concurrently with platelet activation by the LPA content of oxLDL because both pathways are mutually independent. 17 Platelet activation by oxLDL is also independent of apoER2', which is the receptor through which nLDL increases the sensitivity to platelet activating agents. 13 Because under physiological conditions LDL oxidation is probably far from complete, oxLDL particles will change platelet behavior through at least 3 independent signaling routes which are initiated by apoER2' (nLDL), LPA 1 /LPA 3 receptors (LPA in oxLDL), and CD36/SR-A (oxLDL).

Earlier studies showed that the ability of LDL to act as a platelet agonist increases on oxidation, 23–25 and involves LPA-dependent 3,11,12,17 and LPA-independent 17 signaling. LPA-independent signaling was mediated through p38 MAPK activation, which increased with increasing oxidation of LDL. Because p38 MAPK is also an intermediate in nLDL-induced platelet activation, a simple explanation emerged in which nLDL activates a receptor and oxidation enhances this process. The discovery of apoER2' as the exclusive receptor for platelet activation by nLDL made it possible to study whether this receptor recognizes oxLDL. The present data showed that nLDL and partially oxidized LDL shared a similar activation of apoER2'. In contrast, both preparations induced different degrees of p38 MAPK activation with partially oxidized LDL inducing more activation at lower concentrations than nLDL. These data are best explained by assuming that LDL oxidized for <60% contains an nLDL fraction responsible for apoER2' activation with concurrent p38 MAPK phosphorylation and an oxidized fraction that triggers more p38 MAPK activation through a different receptor. This second pathway is unaffected by RAP which blocks ligand binding to LDL receptor family members, chondroitinase ABC which interferes with the binding of nLDL to apoER2' mediated by proteoglycans, and L-NASPA, which blocks ligand binding to LPA receptors. It is therefore independent of LDL receptor family members or LPA receptors. In addition, this pathway remains intact after prolonged incubation with nLDL offering a means to clarify oxLDL signaling without interference of the apoER2' pathway.

Many receptors become inaccessible to ligand binding after prolonged ligand exposure. Examples are glycoprotein Ib and PAR-1 which are internalized after platelet activation and sorted to lysosomes rather than recycled to the plasma membrane. 26,27 After 30 minutes incubation with nLDL, activation of apoER2' was fully reversible showing no signs of irreversible ligand binding, receptor inactivation, or internalization. Surprisingly, this treatment almost completely abolished a second activation of p38 MAPK by nLDL revealing a block between apoER2' activation and p38 MAPK activation generated on prolonged nLDL contact. The cause of this blockade is unclear but may involve mechanisms known to downregulate p38 MAPK such as the inhibitory receptor PECAM-1. 28

OxLDL binds to nucleated cells through SRs expressed on their cell surface. SRs are glycoproteins that recognize a broad variety of ligands, such as oxidized- and glycosylated lipoproteins, anionic phospholipids, apoptotic cells, and fatty acids. 19 SRs identified on platelets so far include CD36, 29 SR-BI, 30 CD68, 31 and LOX-1. 32 Earlier reports described that CuSO 4 -oxidized LDL binds to platelets and that binding to CD36 accounted for about 75% of total binding. 21,33 Binding was disturbed by maleylated human serum albumin, which inhibits ligand binding to SRs of classes A and B. Other receptors might also contribute to oxLDL binding to platelets such as LOX-1, the function of which appears to be restricted to activated platelets. 32 The present study shows that p38 MAPK phosphorylation induced by oxLDL was unaffected by separate addition of inhibitors of CD36 and SR-A. A combination of inhibitors to both scavenger receptors effectively reduced oxLDL-induced activation of p38 MAPK to the level found with nLDL. Consistent with these findings were observations in mice deficient in either CD36 or SR-A that responded normally to oxLDL, but the absence of both receptors either by genetic targeting or addition of inhibitors blocked signaling by oxLDL. These findings indicate that the LPA-independent platelet activation by oxLDL is mediated through the combined involvement of CD36 and SR-A.

In platelet suspensions, oxLDL binding to CD36 interferes with fibrinogen binding to integrin IIb β 3 thereby inhibiting TRAP-induced aggregation. 17 The present study shows that oxLDL increases platelet adhesion to immobilized fibrinogen under flow and that this effect is independent of ADP secretion. Binding of soluble fibrinogen is known to depend on prior activation of IIb β 3, but surface-coated fibrinogen binds to the closed conformation present on resting platelets. 34 At the low shear rate used in the present study (300 s –1 ), binding to immobilized fibrinogen exclusively involves IIb β 3. 35,36 Hence, binding of oxLDL to CD36 might interfere with the activation of IIb β 3 thereby interfering with binding of soluble but not with surface-coated fibrinogen. There is a 63% reduction of the stimulation by oxLDL in the presence of inhibitors of CD36 and SR-A. A similar inhibition is seen at the level of p38 MAPK activation. This suggests that the better adhesion induced by oxLDL is the result of combined activation of CD36 and SR-A and enhanced signaling to p38 MAPK, which is an upstream regulator of thromboxane A 2 formation. 37

Acknowledgments

We thank M. Bezemer (Department of Clinical Chemistry and Haematology, University Medical Center Utrecht, the Netherlands) and R.B. Hildebrand (Division of Biopharmaceutics, Leiden Amsterdam Center for Drug Research, Gorlaeus Laboratories, Leiden University, the Netherlands) for their technical assistance.

Sources of Funding

This study was supported in part by grants 1999B061 (to S.J.A.K.), 2001T041 (to M.V.E.), and 2003B134 (to R.O.) of the Netherlands Heart Foundation, and the Netherlands Thrombosis Foundation (to J.-W.N.A.).

Disclosures

None

【参考文献】
  Steinberg D. Lewis A. Conner Memorial Lecture. Oxidative modification of LDL and atherogenesis. Circulation. 1997; 95: 1062–1071.

Ross R. Atherosclerosis-an inflammatory disease. N Engl J Med. 1999; 340: 115–126.

Siess W, KJZangl, MEssler, MBauer, RBrandl, CCorrinth, RBittman, Tigyi G, Aepfelbacher M. Lysophosphatid acid mediates the rapid activation of platelets and endothelial cells by mildly oxidized low density lipoprotein and accumulates in human atherosclerotic lesions. Proc Natl Acad Sci U S A. 1999; 96: 6931–6936.

Nishi K, Itabe H, Uno M, Kitazato KT, Horiguchi H, Shinno K, Nagahiro S. Oxidized LDL in carotid plaques and plasma associates with plaque instability. Arterioscler Thromb Vasc Biol. 2002; 22: 1649–1654.

Yla-Herttuala S, Palinski W, Butler SW, Picard S, Steinberg D, Witztum JL. Rabbit and human atherosclerotic lesions contain IgG that recognizes epitopes of oxidized LDL. Arterioscler Thromb. 1994; 14: 32–40.

Shaw PX, Horkko S, Tsimikas S, Chang MK, Palinski W, Silverman GJ, Chen PP, Witztum JL. Human-derived anti-oxidized LDL autoantibody blocks uptake of oxidized LDL by macrophages and localizes to atherosclerotic lesions in vivo. Arterioscler Thromb Vasc Biol. 2001; 21: 1333–1339.

Esterbauer H, Gebicki J, Puhl H, Jürgens G. The role of lipid peroxidation and antioxidants in oxidative modifications of LDL. Free Rad Biol Med. 1992; 13: 341–390.

Esterbauer H, Jurgens G, Quehenberger O, Koller E. Autoxidation of human low density lipoprotein: loss of polyunsaturated fatty acids and vitamin E and generation of aldehydes. J Lipid Res. 1987; 28: 495–509.

Steinbrecher UP. Oxidation of human low density lipoprotein results in derivatization of lysine residues of apolipoprotein B by lipid peroxide decomposition products. J Biol Chem. 1987; 262: 3603–3608.

Rother E, Brandl R, Baker DL, Goyal P, Gebhard H, Tigyi G, Siess W. Subtype-selective antagonists of lysophosphatidic Acid receptors inhibit platelet activation triggered by the lipid core of atherosclerotic plaques. Circulation. 2003; 108: 741–747.

Retzer M, Essler M. Lysophosphatidic acid-induced platelet shape change proceeds via Rho/Rho kinase-mediated myosin light-chain and moesin phosphorylation. Cell Signal. 2000; 12: 645–648.

Maschberger P, Bauer M, Baumann-Siemons J, Zangl KJ, Negrescu EV, Reininger AJ, Siess W. Mildly oxidized low density lipoprotein rapidly stimulates via activation of the lysophosphatidic acid receptor Src family and Syk tyrosine kinases and Ca 2+ influx in human platelets. J Biol Chem. 2000; 275: 19159–19166.

Korporaal SJA, Relou IAM, van Eck M, Strasser V, Bezemer M, Gorter G, van Berkel ThJC, Nimpf J, Akkerman JWN, Lenting PJ. Binding of low-density lipoprotein to platelet apolipoprotein E receptor 2' results in phosphorylation of p38 MAPK. J Biol Chem. 2004; 279: 52526–52534.

Kim DH, Iijima H, Goto K, Sakai J, Ishii H, Kim HJ, Suzuki H, Kondo H, Saeki S, Yamamoto T. Human apolipoprotein E receptor 2. A novel lipoprotein receptor of the low density lipoprotein receptor family predominantly expressed in brain. J Biol Chem. 1996; 271: 8373–8380.

Riddell DR, Vinogradov DV, Stannard AK, Chadwick N, Owen JS. Identification and characterization of LRP8 (apoER2) in human blood platelets. J Lipid Res. 1999; 40: 1925–1930.

Hackeng CM, Relou IAM, Pladet MW, Gorter G, van Rijn HJM, Akkerman JWN. Early platelet activation by low density lipoprotein via p38MAP kinase. Thromb Haemost. 1999; 82: 1749–1756.

Korporaal SJA, Gorter G, van Rijn HJM, Akkerman JWN. The effect of oxidation on the platelet-activating properties of low density lipoprotein. Arterioscler Thromb Vasc Biol. 2005; 25: 867–872.

Boren J, Olin K, Lee I, Chait A, Wight TN, Innerarity TL. Identification of the principal proteoglycan-binding site in LDL. A single-point mutation in apo-B100 severely affects proteoglycan interaction without affecting LDL receptor binding. J Clin Invest. 1998; 101: 2658–2664.

Greaves DR, Gough PJ, Gordon S. Recent progress in defining the role of scavenger receptors in lipid transport, atherosclerosis and host defence. Curr Opin Lipidol. 1998; 9: 425–432.

Puente Navazo MD, Daviet L, Ninio E, McGregor JL. Identification on human CD36 of a domain (155-183) implicated in binding oxidized low-density lipoproteins (Ox-LDL). Arterioscler Thromb Vasc Biol. 1996; 16: 1033–1039.

Volf I, Moeslinger T, Cooper J, Schmid W, Koller E. Human platelets exclusively bind oxidized low density lipoprotein showing no specificity for acetylated low density lipoprotein. FEBS Lett. 2000; 449: 141–145.

Remijn JA, Wu JP, Jeninga EH, Ijsseldijk M, van Willigen G, de Groot PG, Sixma JJ, Nurden AT, Nurden P. Role of ADP receptor P2Y(12) in platelet adhesion and thrombus formation in flowing blood. Arterioscler Thromb Vasc Biol. 2002; 22: 686–691.

Weidtmann A, Scheithe R, Hrboticky N, Pietsch A, Lorenz R, Siess W. Mildly oxidized LDL induces platelet aggregation through activation of phospholipase A 2. Arterioscler Thromb Vasc Biol. 1995; 15: 1131–1138.

Ardlie NG, Selley ML, Simons LA. Platelet activation by oxidatively modified low density lipoproteins. Atherosclerosis. 1989; 76: 117–124.

Chen LY, Mehta P, Mehta JL. Oxidized LDL decreases L-arginine uptake and nitric oxide synthase protein expression in human platelets. Relevance of the effect of oxidized LDL on platelet function. Circulation. 1996; 93: 1740–1746.

Nurden P. Bidirectional trafficking of membrane glycoproteins following platelet activation in suspension. Thromb Haemost. 1997; 78: 1305–1315.

Coughlin SR. How the protease thrombin talks to cells. Proc Natl Acad Sci U S A. 1999; 96: 11023–11027.

Relou IAM, Gorter G, Ferreira IA, van Rijn HJM, Akkerman JWN. Platelet endothelial cell adhesion molecule-1 (PECAM-1) inhibits low density lipoprotein-induced signaling in platelets. J Biol Chem. 2003; 278: 32638–32644.

Tandon NN, Lipsky RH, Burgess WH, Jamieson GA. Isolation and characterization of platelet glycoprotein IV (CD36). J Biol Chem. 1989; 264: 7570–7575.

Imachi H, Murao K, Cao W, Tada S, Taminato T, Wong NC, Takahara J, Ishida T. Expression of human scavenger receptor b1 on and in human platelets. Arterioscler Thromb Vasc Biol. 2003; 23: 898–904.

Walker G, Bourguignon LY. Membrane-associated 41-kDa GTP-binding protein in collagen-induced platelet activation. FASEB J. 1990; 4: 2925–2933.

Chen M, Kakutani M, Naruko T, Ueda M, Narumiya S, Masaki T, Sawamura T. Activation-dependent surface expression of LOX-1 in human platelets. Biochem Biophys Res Commun. 2001; 282: 153–158.

Volf I, Roth A, Cooper J, Moeslinger T, Koller E. Hypochlorite modified LDL are a stronger agonist for platelets than copper oxidized LDL. FEBS Lett. 2000; 483: 155–159.

Lindon JN, Kushner L, Salzman EW. Platelet interaction with artificial surfaces: in vitro evaluation. Methods Enzymol. 1989; 169: 104–117.

Endenburg SC, Hantgan RR, Sixma JJ, de Groot PG, Zwaginga JJ. Platelet adhesion to fibrin(ogen). Blood Coagul Fibrinolysis. 1993; 4: 139–142.

Savage B, Saldivar E, Ruggeri ZM. Initiation of platelet adhesion by arrest onto fibrinogen or translocation on von Willebrand factor. Cell. 1996; 84: 289–297.

Kramer RM, Roberts EF, Um SL, Borsch-Haubold AG, Watson SP, Fisher MJ, Jakubowski JA. p38 mitogen-activated protein kinase phosphorylates cytosolic phospholipase A2 (cPLA2) in thrombin-stimulated platelets. Evidence that proline-directed phosphorylation is not required for mobilization of arachidonic acid by cPLA2. J Biol Chem. 1996; 271: 27723–27729.

作者单位：Department of Clinical Chemistry and Haematology (S.J.A.K., J.A., M.I., T.L., P.J.L., J.-W.N.A.), University Medical Center Utrecht, and Institute of Biomembranes, Utrecht University, the Netherlands; and the Division of Biopharmaceutics (S.J.A.K., M.V.E., R.O., T.J.C.V.B.), Leiden/Amsterdam Center