清道夫受体,氧化低密度脂蛋白和动脉粥样硬化

Scavenger Receptors, Oxidized LDL, and Atherosclerosis

AGNES BOULLIER, DAVID A. BIRD, MI-KYUNG CHANG, EDWARD A. DENNIS, PETER FRIEDMAN, KRISTIN GILLOTTE-TAYLOR, SOHVI H?RKK?,

WULF PALINSKI, OSWALD QUEHENBERGER, PETER SHAW,

DANIEL STEINBERG, VALESKA TERPSTRA, AND JOSEPH L. WITZTUM Department of Medicine, University of California, San Diego, La Jolla,

California 92093, USA

A BSTRACT: Oxidized LDL (OxLDL) competes with oxidatively damaged and

apoptotic cells for binding to mouse peritoneal macrophages, implying the

presence of one or more common domains. However, the nature of the ligands

involved has not been determined. Studies in this laboratory over the last sev-

eral years provide evidence that oxidized phospholipids, present in OxLDL and

also in the membrane of apoptotic cells, represent one such ligand. These oxi-

dized phospholipids, either in the lipid phase of OxLDL or becoming attached

covalently to apoprotein B during LDL oxidation, have been shown to play a

major role in the binding of OxLDL to CD36 and to SR-B1 expressed in trans-

fected cells. The lipid and protein moieties compete with each other to some ex-

tent, indicating that they are binding to at least one common site. A monoclonal

antibody selected because of its reactivity with OxLDL proved to be an anti-

body against oxidized phospholipids (but not native phospholipids). This anti-

body (EO6) blocked the uptake of OxLDL by CD36 and by SR-B1 in

transfected cells by as much as 80%; it also inhibited macrophage phagocytosis

of apoptotic cells by about 40%. Thus, the persistence of receptors for OxLDL

during evolution is probably accounted for by their role in recognition of

ligands on the surfaces of oxidatively damaged or apoptotic cells. This has im-

portant implications in biology generally and specifically in atherogenesis, be-

cause apoptosis is a prominent feature of late lesions.

K EYWORDS: scavenger receptors; oxidized LDL; atherosclerosis; apoptotic

cells

INTRODUCTION

Most of the lipid-laden “foam cells” in early atherosclerotic lesions represent monocyte/macrophages that have taken up lipoproteins in the subendothelial space. This is not the result of the uptake of native LDL, which cannot induce cholesterol accumulation in monocyte/macrophages, but rather it is due to the uptake of one or more modified forms of LDL. The best studied modification that can cause choles-terol accumulation is oxidative modification. The large body of evidence implicating Address for correspondence: Daniel Steinberg, M.D., Ph.D., Department of Medicine, Univer-sity of California San Diego, La Jolla, CA 92093-0682. Voice: 858-534-0569; fax: 858-534-2005.

dsteinberg@https://www.360docs.net/doc/a24932909.html,

214

215 BOULLIER et al.: SCAVENGER RECEPTORS, LDL, AND ATHEROSCLEROSIS

oxidized LDL in atherogenesis has been reviewed elsewhere.1 Oxidized LDL (Ox-LDL) binds with high affinity to several receptors on the macrophage plasma mem-brane, receptors that can internalize OxLDL and lead to its degradation. These include scavenger receptor A, the first OxLDL receptor to be characterized and cloned,2 CD36,3 CD68,4 LOX-1,5 and possibly others. The quantitative importance of some of these OxLDL receptors in vivo has not been directly assessed, but it is clear that SRA and CD36 are quantitatively important. In knockout experiments, the deletion of either one of these two receptors results in a significant amelioration of the severity of experimental atherosclerosis in apoE-deficient mice.6,7 Those find-ings are consonant with the interpretation that uptake of OxLDL leading to foam cell formation is an important process in atherogenesis. However, other interpretations are possible because these receptors have a very broad specificity (so-called “pattern recognition receptors”), and loss of function could influence atherogenesis in other ways as well.

WHAT ARE THE “NATURAL LIGANDS” FOR MACROPHAGE

SCAVENGER RECEPTORS?

For years, our group at the Specialized Center of Research on Molecular Medi-cine and Atherosclerosis in La Jolla has concerned itself with the question: Why have scavenger receptors persisted in evolution? They are found in all mammalian species, and homologues have been described all the way back to sea urchins.8 Since the uptake of OxLDL is proatherogenic, there would be no selective advantage in ex-pressing these receptors; in any case, atherosclerosis does not affect individuals until after the childbearing period. Therefore, the scavenger receptors must have been se-lected for some function other than their ability to recognize OxLDL. Presumably, there are natural ligands for these receptors that are similar in structure to one or more components of OxLDL. We speculated that oxidative damage to plasma mem-branes of cells might generate products analogous to those generated by oxidative damage to LDL.

The first test of this hypothesis was to determine whether OxLDL could compete with oxidatively damaged cells for binding to macrophage scavenger receptors. We showed that indeed the binding and phagocytosis of oxidatively damaged red blood cells or of apoptotic thymocytes to mouse peritoneal macrophages were very strong-ly inhibited by OxLDL but not by native LDL or acetyl LDL.9,10 We then went on to ask whether the binding of OxLDL was attributable to modifications of the protein moiety, the lipid moieties, or both.11 We first extracted the lipids from OxLDL ex-haustively using chloroform-methanol, reconstituted these lipids into a microemul-sion containing a lipid-soluble fluorescent marker, and studied their binding to peritoneal macrophages. We found that there was indeed specific, saturable binding and that these microemulsions prepared from the lipids of OxLDL competed very effectively for the binding of intact OxLDL to the macrophage. Lipids prepared from native LDL in a similar fashion were without effect.

The protein moiety of OxLDL after lipid extraction can be resolubilized in octyl-glucoside.12 This is not the case for native LDL: intact apoB is virtually insoluble, whereas the apoB in OxLDL, having been extensively degraded, is readily solubi-

216ANNALS NEW YORK ACADEMY OF SCIENCES

lized. The radioiodinated apoB prepared from OxLDL binds in a saturable fashion to macrophages and can compete with intact OxLDL for binding.

At this point we could not rule out the possibility that the lipid moiety and the protein moiety were actually binding to different receptors and competing with in-tact OxLDL, because it also was binding to different receptors and being recognized by some of them on the basis of the modified protein and by others on the basis of the modified lipid. However, we suspected that the lipid and the protein moieties must be, to some extent, binding to a common ligand-binding site, because in some experiments the extent of inhibition of intact OxLDL by each of the two moieties ap-proached 80%.11 Direct evidence supporting this came from the demonstration that the microemulsions of OxLDL lipids could compete for the binding of OxLDL apoB and vice versa.13 Finally, as schematized in F IGURE 1 and discussed below, we have shown that specific binding of intact OxLDL to CD36-transfected cells is inhibited significantly by both the OxLDL lipids and the resolubilized OxLDL apoB. More-over, the separated lipid and protein moieties partially inhibit each other ’s specific binding.14

FIGURE 1.Schematic representation of how intact oxidized LDL (OxLDL) and its component apoprotein and lipid moieties interact with CD36 and with each other. Intact Ox-LDL (left ), the apoprotein moiety generated by extraction with chloroform-methanol and resolubilized in octylglucoside (middle ), and the lipid moiety reconstituted into a micro-emulsion (right ) all bind in a saturable manner to CD36. The specific binding of the intact OxLDL can be partially inhibited by either the protein moiety or the lipid moiety. Finally,the apoprotein and the lipids can compete with each other for binding, because the “protein ”moiety actually contains oxidized phospholipids covalently bound to it.

217 BOULLIER et al.: SCAVENGER RECEPTORS, LDL, AND ATHEROSCLEROSIS RECOGNITION OF O X LDL VIA OXIDIZED PHOSPHOLIPIDS

Modification of the phospholipids seemed the most likely basis for receptor rec-ognition of OxLDL. The phospholipids constitute by far the largest majority of lipid on the surface of the LDL particle, whereas the neutral lipids are buried in the core. We subjected pure 1-palmitoyl-2-arachidonoyl-phosphatidylcholine (PAPC) to air oxidation at room temperature for varying lengths of time and tested its ability to compete with OxLDL either in the form of a liposome (made up with nonoxidized phosphatidylcholine and cholesterol) or as a conjugate with one or another protein. The oxidized PAPC, in the form of a liposome containing cholesterol and phosphati-dylcholine, was an effective inhibitor.13 Watson et al.15 had previously characterized some of the biologically active phospholipids in minimally oxidized LDL and iden-tified one of them as 1-palmitoyl-2-(5-oxovaleryl) phosphatidylcholine (POVPC). We therefore synthesized POVPC and showed that liposomes containing this fully characterized, pure oxidized phospholipid were able to do exactly what the mixed oxidation products from PAPC were able to do. Moreover, POVPC covalently linked to bovine serum albumin was also an effective competitor for OxLDL.16

OXIDIZED PHOSPHOLIPIDS BIND COVALENTLY TO APOB IN O X LDL

Now it was possible to explain how both the apoprotein moiety and the lipid moi-ety could compete for the binding of OxLDL and how the lipid moiety and the apo-protein moiety could compete with each other for binding. As schematized in F IGURE 1, we proposed that during the oxidation of LDL, partially oxidized phospholipids with an active aldehyde group at the omega end of the fatty acid in the two position became covalently attached to the proteins via their epsilon lysine amino groups or other available free amino groups. This covalently linked phospholipid would, of course, not be removed even by rigorous chloroform-methanol extraction and so what we designate the “protein moiety” (i.e., what remains after lipid extraction) still contains oxidized phospholipids covalently linked to the protein. That hypothesis has now been firmly established in two ways.

First, a monoclonal antibody that specifically recognizes oxidized phospho-lipid,17 antibody EO6, was shown to recognize intact OxLDL and the isolated apoB from OxLDL and the isolated lipid moiety from OxLDL.13,16 Moreover, this same monoclonal antibody inhibited macrophage binding and uptake of both intact Ox-LDL and of the isolated protein and lipid moieties. More recently, we studied the binding of OxLDL and of its component parts to specific scavenger receptors —mouse CD 3614 and mouse SR-B118— expressed in transfected cells. The results were very similar to those obtained with the mouse peritoneal macrophage, that is, either the reconstituted microemulsion or the apoprotein moiety of OxLDL could compete effectively for the binding of intact OxLDL. As with the intact peritoneal macrophages, the binding of OxLDL to the transfected cells was inhibited by mon-oclonal antibody EO6, which recognizes oxidized phospholipids, and by the POV-PC-albumin adduct.

The second line of evidence for covalent binding of phospholipid to apoB in Ox-LDL came from direct chemical analysis. OxLDL was subjected to exhaustive lipid extraction, and the residual protein was digested for determination of any remaining

218ANNALS NEW YORK ACADEMY OF SCIENCES phosphate. When this was done with native LDL, almost no detectable phosphate re-mained with the apoprotein. However, when the same procedure was carried out with OxLDL, an average of 70 moles of phosphate remained attached to each mole of apoB. As already mentioned, the delipidated apoprotein from OxLDL is recognized by monoclonal antibody EO6 and that was confirmed in these studies. However, if the apoprotein was first subjected to mild alkaline hydrolysis, then it no longer re-acted with the antibody. Analysis of the material released by the mild alkaline hy-drolysis showed that it contained saturated fatty acid, phosphate, and choline in an approximately 1:1:1 molar ratio, that is, the hydrolysis of the attached phospholipid at its ester bonds destroyed the antigenic potential of the oxidized phospholipid. Pre-sumably, the residual, partially degraded fatty acid chain in the two position re-mained attached to the protein amino groups.

Our findings establish that a significant fraction of the binding of intact OxLDL and of OxLDL apoB to macrophage scavenger receptors is due to oxidized phospho-lipids, but other modifications are undoubtedly involved, such as generation of malondialdehyde and 4-hydroxynonenal and their binding to apoB.

EVIDENCE THAT BINDING OF APOPTOTIC CELLS BY SCAVENGER RECEPTORS IS IN PART DUE TO RECOGNITION OF OXIDIZED

PHOSPHOLIPIDS OR HOMOLOGOUS STRUCTURES IN THE

PLASMA MEMBRANE

As just discussed, OxLDL can compete with oxidatively damaged cells and apoptotic cells for binding to macrophage scavenger receptors, and so they must have one or more ligands in common. The availability of antibody EO6 provided an ideal tool for testing whether that common ligand was oxidized phospholipid.20 Apoptotic thymocytes or endothelial cells were prepared either by incubating the cells with dexamethasone or by serum deprivation, respectively. It was shown that EO6 reacted with epitopes on the membranes of these apoptotic cells, but it did not recognize viable cells. Moreover, antibody EO6 could inhibit the phagocytosis of apoptotic thymocytes by elicited peritoneal macrophages, as schematized in F IGURE2.

We recently showed that apoptotic cells contain oxidized phospholipids that are recognized by antibody EO6 (Chang et al., unpublished results). However, we still have not defined the precise chemical nature of the oxidized phospholipids (or oxi-dized phospholipid-protein complexes) on the surface of apoptotic cells that are rec-ognized by scavenger receptors. The antibody recognizes several species of oxidized phosphatidylcholine in addition to POVPC (Friedman et al., unpublished results). A large body of evidence strongly supports the possibility that clustering of phosphati-dylserine (PS) residues on the surface of apoptotic cells is an important ligand that mediates their recognition by macrophages.21–23 What we are proposing is that ox-idized phospholipids, generated during apoptosis or as cells go on to die, also play a role. Our current data suggest that in response to an apoptotic stimulus the appear-ance of PS on the cell surface is an early event, whereas the appearance of the oxi-dation-specific epitopes recognized by EO6 begins shortly thereafter (Chang et al., unpublished data).

219

BOULLIER et al .: SCAVENGER RECEPTORS, LDL, AND ATHEROSCLEROSIS THE SPECIAL NATURE OF EO6

Interpretation of these results must take into account some recent findings regard-ing antibody EO6. Shaw et al.24 showed that the DNA sequence of the variable re-gion of EO6 is 100% identical to that of a naturally occurring germ-line antibody designated T15. EO6 and T15 are structurally and functionally equivalent. The T15antibody clan has been extensively studied for about 30 years and has been shown to recognize phosphorylcholine (choline phosphate, the head group of phosphatidyl-choline), which is a common constituent of the cell wall of many pathogens. T15 and related antibodies protect mice against lethal doses of Streptococcus pneumoniae .However, mice produce this antibody even under germ-free conditions, and the na-ture of the inducing antigen has been obscure. In the light of the work just discussed,it is now suggested that oxidatively modified phospholipids on the surface of apop-totic cells (or equivalent structures found in OxLDL) may represent one of the en-

dogenous antigens that modulate the level of expression of T15/EO6.

FIGURE 2.Schematic diagram to indicate that the recognition of apoptotic cells and of OxLDL by macrophage scavenger receptors is in part attributable to structures common to both of them, namely, oxidized phospholipids. Monoclonal antibody EO6, as discussed in the text, is directed against oxidized phospholipids. POVPC is an oxidized phospholipid with palmitic acid esterified to the glycerol backbone at the 1 position and 5-oxovaleric acid esterified to the glycerol backbone at the 2 position. The complex of POVPC with bovine serum albumin (BSA) by Schiff base formation with lysine epsilon amino groups is an ef-fective inhibitor of both the binding of apoptotic cells and of OxLDL.

220ANNALS NEW YORK ACADEMY OF SCIENCES

SUMMARY

Scavenger receptors, at least some of them, have been selected during evolution because they can participate in the recognition and clearance of bacteria and apop-totic cells. We suggest that the damaged cells are in part recognized by virtue of the presence of oxidatively modified phospholipids on their surface. Oxidation of LDL generates modified phospholipids of the same or similar structure, and OxLDL is therefore also recognized by this class of scavenger receptors and by some natural antibodies against oxidized phospholipids. Macrophages and natural antibodies are both components of the innate immune system. Our studies show that oxidized phos-pholipids are common ligands for both scavenger receptors and some important natural antibodies.

An obvious proposition is that inhibition of scavenger receptor function, with a decrease in the uptake of oxidized LDL, might be antiatherogenic in humans, as in the mouse knockouts already reported. However, we should add a caveat. Because these receptors also play a role in protection against microbial disease and in the clearance of apoptotic cells, interference with their function must be titrated or tar-geted selectively to arterial wall to avoid unacceptable side effects. For example, even though knockouts of SRA or CD36 inhibit generation of early lesions, a defect in clearance of apoptotic, lipid-laden macrophages could be a factor in the formation of the necrotic lipid core that characterizes the vulnerable plaque, as schematized in F IGURE 3. It should also be noted that the presence of OxLDL in the lesion would tend to inhibit scavenging of the apoptotic cells, because they would be competing for the same receptors. This may become a significant factor especially in the later lesions.

FIGURE3.Schema to indicate how failure to clear apoptotic cells before they become necrotic may favor the formation of a lesion with a large pool of lipid, that is, a vulnerable plaque.

BOULLIER et al.: SCAVENGER RECEPTORS, LDL, AND ATHEROSCLEROSIS

221

ACKNOWLEDGMENTS

The original studies cited were supported by National Heart, Lung and Blood In-stitute Grants HL56989 (Specialized Center of Research in Molecular Medicine and Atherosclerosis) and HL57505. We also acknowledge valuable support from the American Heart Association and the California Tobacco-Related Disease Program.

REFERENCES

1.S TEINBERG, D. & J.L. W ITZTUM. 1999. Lipoproteins, lipoprotein oxidation, and athero-

genesis. In Molecular Basis of Cardiovascular Disease. K.E Chien, ed. :428–476.

W.B. Saunders. Philadelphia.

2.K ODAMA, T., P. R EDDY, C. K ISHIMOTO, et al. 1988. Purification and characterization of

a bovine acetyl low density lipoprotein receptor. Proc. Natl. Acad. Sci. USA 85:

9238–9242.

3.E NDEMANN, G., L.W. S TANTON, K.S. M ADDEN, et al. 1993. CD36 is a receptor for oxi-

dized low density lipoprotein. J. Biol. Chem. 268: 11811–11816.

4.R AMPRASAD, M.P., W. F ISCHER, J.L. W ITZTUM, et al. 199

5. The 94- to 97-kDa mouse

macrophage membrane protein that recognizes oxidized low density lipoprotein and phosphatidylserine-rich liposomes is identical to macrosialin, the mouse homologue of human CD68. Proc. Natl. Acad. Sci. USA 92: 9580–9584.

5.S AWAMURA, T., N. K UME, T. A OYAMA, et al. 1997. An endothelial receptor for oxidized

low-density lipoprotein. Nature 386: 73–77.

6.S UZUKI, H., Y. K URIHARA, M. T AKEYA, et al. 199

7. A role for macrophage scavenger

receptors in atherosclerosis and susceptibility to infection. Nature 386: 292–296. 7.F EBBRAIO, M., E.A. P ODREZ, J.D. S MITH, et al. 2000. Targeted disruption of the class B

scavenger receptor CD36 protects against atherosclerotic lesion development in mice. J. Clin. Invest 105: 1049–1056.

8.A BRAMS, J.M., A. L UX, H. S TELLER & M. K RIEGER. 1992. Macrophages in Drosophila

embryos and L2 cells exhibit scavenger receptor-mediated endocytosis. Proc. Natl.

Acad. Sci. USA 89: 10375–10379.

9.S AMBRANO, G.R. & D. S TEINBERG. 1994. Recognition of oxidatively damaged erythro-

cytes by a macrophage receptor with specificity for oxidized low density lipoprotein.

Proc. Natl. Acad. Sci. USA 91: 3265–3269.

10.S AMBRANO, G.R. & D. S TEINBERG. 1995. Recognition of oxidatively damaged and

apoptotic cells by an oxidized low density lipoprotein receptor on mouse peritoneal macrophages: role of membrane phosphatidylserine. Proc. Natl. Acad. Sci. USA 92: 1396–1400.

11.T ERPSTRA, V., D.A. B IRD & D. S TEINBERG. 1998. Evidence that the lipid moiety of oxi-

dized low density lipoprotein plays a role in its interaction with macrophage recep-tors. Proc. Natl. Acad. Sci. USA 95: 1806–1811.

12.P ARTHASARATHY, S., L.G. F ONG, D. O TERO, et al. 1987. Recognition of solubilized

apoproteins from delipidated, oxidized low density lipoprotein (LDL) by the acetyl-LDL receptor. Proc. Natl. Acad. Sci. USA 84: 537–540.

13.B IRD, D.A., K.L. G ILLOTTE, S. H?RKK?, et al. 1999. Receptors for oxidized low-den-

sity lipoprotein on elicited mouse peritoneal macrophages can recognize both the modified lipid moieties and the modified protein moieties: implications with respect to macrophage recognition of apoptotic cells. Proc. Natl. Acad. Sci. USA 96: 6347–6352.

14.B OULLIER, A., K.L. G ILLOTTE, S. H?RKK?, et al. 2000. The binding of oxidized low

density lipoprotein to mouse CD36 is mediated in part by oxidized phospholipids that are associated with both the lipid and protein moieties of the lipoprotein. J. Biol.

Chem. 275: 9163–9169.

15.W ATSON,A.D., N. L EITINGER, M. N AVAB, et al. 1997. Structural identification by mass

spectrometry of oxidized phospholipids in minimally oxidized low density lipopro-

222ANNALS NEW YORK ACADEMY OF SCIENCES

tein that induce monocyte/endothelial interactions and evidence for their presence in vivo. J. Biol. Chem. 272: 13597–13607.

16.H?RKK?, S., D.A. B IRD, E. M ILLER, et al. 1999. Monoclonal autoantibodies specific for

oxidized phospholipids or oxidized phospholipid-protein adducts inhibit macrophage uptake of oxidized low-density lipoproteins. J. Clin. Invest 103: 117–128.

17.P ALINSKI, W., S. H?RKK?, E. M ILLER, et al. 1996. Cloning of monoclonal autoantibod-

ies to epitopes of oxidized lipoproteins from apolipoprotein E-deficient mice. Dem-onstration of epitopes of oxidized low density lipoprotein in human plasma. J. Clin.

Invest 98: 800–814.

18.G ILLOTTE-T AYLOR, K., A. B OULLIER, J.L. W ITZTUM, et al. 2001. Scavenger receptor

class B type I as a receptor for oxidized low density lipoprotein. J. Lipid Res. 42: 1474–1482.

19.G ILLOTTE, K.L., S. H?RKK?, J.L. W ITZTUM, et al. 2000. Oxidized phospholipids, linked

to apolipoprotein B of oxidized LDL, are ligands for macrophage scavenger recep-tors. J. Lipid Res. 41: 824–833.

20.C HANG, M.K., C. B ERGMARK, A. L AURILA, et al. 1999. Monoclonal antibodies against

oxidized low-density lipoprotein bind to apoptotic cells and inhibit their phagocyto-sis by elicited macrophages: evidence that oxidation-specific epitopes mediate mac-rophage recognition. Proc. Natl. Acad. Sci. USA 96: 6353–6358.

21.S AVILL, J., V. F ADOK, P. H ENSON, et al. 1993. Phagocyte recognition of cells undergo-

ing apoptosis. Immunol. Today 14: 131–136.

22.F ADOK,V.A., D.L. B RATTON, D.M. R OSE, et al. 2000. A receptor for phosphati-

dylserine-specific clearance of apoptotic cells. Nature 405: 85–90.

23.F ADOK, V.A., M.L. W ARNER, D.L. B RATTON, et al. 1998. CD36 is required for phagocy-

tosis of apoptotic cells by human macrophages that use either a phosphatidylserine receptor or the vitronectin receptor (alpha v beta 3). J. Immunol. 161: 6250–6257. 24.S HAW, P.X., S. H?RKK?, M.K. C HANG, et al. 2000. Natural antibodies with the T15

idiotype may act in atherosclerosis, apoptotic clearance, and protective immunity. J.

Clin. Invest. 105: 1731–1740.

DISCUSSION

J. S MITH (Rockefeller University, New York, New York, USA): In the CD36 knock-out mouse there is no evidence of apoptotic cells piling up anywhere. Is this the case because there are redundant systems for clearing up apoptotic cells? The ABCA1 protein, involved in Tangier disease, has also been implicated in the uptake of apop-totic cells by Chimini’s laboratory in Marseilles.

D ANIEL S TEINBERG: Certainly there is a lot of redundancy. The list of receptors just went up by one. We don’t know how many more there are. It is also true that the CD36 knockout mice have significantly less atherosclerosis. It provides strong evi-dence, along with the Suzuki knockout of SRA, that uptake by way of the scavenger receptors is a proatherogenic process. There was some doubt about whether that would necessarily be true, because oxidized LDL is cytotoxic and taking it up might be a good thing to avoid induction of apoptosis or damage of endothelial cells, but on balance it would seem that taking up that oxidized LDL is proatherogenic and that is somewhat reassuring. It adds strength to the oxidation hypothesis because, as far as I know, the other functions of those receptors would not necessarily be expected to be proatherogenic, whereas the function of taking up oxidized LDL and becoming a foam cell certainly could be a central issue in atherogenesis. The uptake in the SRA knockout was not nearly as dramatic. The fact that the results with nonperitoneal macrophages with all receptors functioning are very similar to the results with CD36

223 BOULLIER et al.: SCAVENGER RECEPTORS, LDL, AND ATHEROSCLEROSIS

is lending support to the contention that Silverstein and Hajjar have made that CD36 is a dominantly important receptor in regard to atherogenesis.

K. H AYASHIDA(Kyoto University, Kyoto, Japan): Long fatty acids also are ligands for CD36. Did you examine competition of long chain fatty acids for apop-totic cells?

S TEINBERG: We did not, and I think that is a very good point. CD36 is by no means a receptor that plays just this role. In fact, it was cloned by one group on the basis of its binding with fatty acid, and it should be studied with regard to the binding of hydroxy-eicosatrienoic acid and other oxidation products from LDL, but we have not done that. I should mention that the elegant work by Silverstein showed that the rat that is deficient in CD36 goes blind because it cannot take up the outer rods as they die, and they accumulate and lead to retinal degeneration. Its role as a receptor for apoptotic cells is independently established by findings such as that.

T. K ITA (Kyoto University, Kyoto, Japan): Have you ever checked the presence of the phospholipid POVPC in apoptotic cells?

S TEINBERG: I have not. Joe Witztum is pursuing that. He knows that the E06 an-tibody reacts, and that is indirect evidence for the presence of the POVPC. Recently, he has collaborated with a group running a mass spectrometer and has evidence that, indeed, POVPC is present in the lipid extract from apoptotic cells.