Immunolocalization of β2-Glycoprotein I (Apolipoprotein H) to Human Atherosclerotic Plaques
Potential Implications for Lesion Progression
Background—β2-Glycoprotein I (β2GPI) is a major antigenic target of antiphospholipid antibodies, which possesses natural anticoagulant properties. The aim of the present study was to determine its presence and localization within human atherosclerotic plaques and to study its association with endothelial cells and monocyte macrophages in vitro.
Methods and Results—Human atherosclerotic lesions were obtained after carotid endarterectomies and studied immunohistochemically with anti-β2GPI as well as antibodies to CD4/CD8, macrophages, and adhesion molecules. In vitro, human umbilical vein endothelial cells (HUVECs) and U937 (myelomonocytic cell line) cells were investigated for their ability to associate with radiolabeled β2GPI. We found β2GPI to be abundantly expressed within the subendothelial regions and intimal-medial borders of human atherosclerotic plaques and to colocalize with CD4-positive lymphocytes. This observation was confirmed by Western blot applied on homogenates of atherosclerotic lesions with anti-β2GPI antibodies. Both HUVECs and U937 cells bound labeled β2GPI, and the process was inhibited by oxidized LDL and not by native LDL.
Conclusions—The abundant presence of human β2GPI within the lesions, its association with endothelial cells and macrophages, and its colocalization with CD4-positive lymphocytes suggests that it may serve as a target for an immune-mediated reaction that can influence lesion progression.
The multifactorial determinants that are involved in the development and progression of atherosclerosis have received increasing attention in recent years.1 Accordingly, the role of autoimmune factors has attracted intensive research aimed at defining the autoantigenic materials expressed within plaques that may influence the fate of lesions.2 Candidate proteins that have been mentioned include modified forms of LDL,3 heat shock proteins (reviewed in Reference 44 ), and β2-glycoprotein I (β2GPI).5
β2GPI is a highly glycosylated plasma protein with an approximate molecular weight of 50 kDa that avidly binds negatively charged surfaces and substances (ie, heparin, DNA, dextran sulfate, anionic phospholipids, and apoptotic cells) (reviewed in Reference 66 ). β2GPI contains 5 short consensus repeat domains, and its partial association with various lipoproteins results in its synonymous designation as apolipoprotein H. The recent interest in β2GPI results from the observation that it serves as a major antigenic target of thrombosis-associated antiphospholipid antibodies (aPL).6
The physiological role of β2GPI is still obscure, yet it possesses several properties that may bear relevance to progression of human atherosclerotic plaques (summarized in Reference 66 ): (1) it binds activated platelets and apoptotic cells (on exposure of inner-membrane phosphatidylserine); (2) it inhibits intrinsic blood coagulation pathway and ADP-dependent platelet aggregation; (3) it serves a requisite role in the activation of endothelial cells induced by aPL; and (4) it may assist in mediating clearance of senescent cells and foreign particles from circulation.
We have recently shown that immunization of LDL receptor–deficient mice with β2GPI results in acceleration of aortic fatty streaks.5 The aortic sinus from the mice was infiltrated by CD4 lymphocytes. In the present study, we observed that β2GPI was present in atherosclerotic plaques, and we defined its relationship with prevailing immunopotent cells. We next reinforced the findings in vitro by demonstrating the interaction of β2GPI with cellular components of human plaques.
Antigens and Antibodies
Antibodies used for immunohistochemistry and Western blotting were as follows: polyclonal mouse and rabbit anti-β2GPI antibodies generated by immunization of the animals with human β2GPI and affinity purified against the respective antigen; monoclonal mouse anti-β2GPI antibodies (Cof-18 and Cof-21), kindly provided by Professor Takao Koike (Sapporo, Japan); anti-macrophage antibodies (CD68, clone PG-M1), anti–VCAM-1 (CD106, clone 1.4C3), and anti–E-selectin (CD62E, clone 1.2B6), all from Dako; and anti-CD4 (clone Z058), anti-CD8 (clone SPV-T8), and anti–ICAM-1 (CD54, clone MY13), all from Zymad.
Samples of human atherosclerotic plaques were obtained from patients after carotid endarterectomy (n=5). Additional frozen sections of human carotid plaques (n=9) and normal carotid arteries (n=5) were obtained after postmortem procedures on humans who died of noncardiac causes.
Immunohistochemical Study of Human Lesions
Immunohistochemical staining for human β2GPI was performed on 5-μm-thick frozen sections of human carotid plaques. After fixation with methanol and acetone, sections were blocked with nonimmune rabbit and goat sera, followed by incubation with CAS blocking reagent. Subsequently, the primary antibody (either of the anti-β2GPI antibodies; 20 μm/mL) was added for 1 hour at room temperature. After washing, biotinylated affinity-purified goat anti-mouse/rabbit antibodies (Jackson) were added. The slides were then incubated with 0.3% H2O2, followed by additional rinses and incubation with streptavidin-peroxidase conjugate (Jackson) for 30 minutes at room temperature. The slides were developed with 3-amino-9 ethylcarbazole substrate (Dako) for 15 minutes and counterstained with hematoxylin. As a control antibody, we used normal mouse/rabbit IgG.
Immunoblotting for Detection of β2GPI in Human Plaques
Fresh human atherosclerotic plaques were obtained within 1 hour after carotid endarterectomy and placed immediately in sterile PBS with 5 U/mL heparin at 4°C. The plaques were separated from adherent surrounding tissues and blood clots, after which the tissues were extensively washed until no contaminating blood was observed. The tissue was then homogenized (polytron) in PBS containing 2 mmol of a protease inhibitor (ie, PMSF; Sigma). The homogenate was stirred for 2 hours at 4°C. After centrifugation, the supernatant was used for detection of the presence of β2GPI. Supernatants of the respective plaques and of nonlesioned arteries were applied on 10% acrylamide SDS gel under reduced conditions, transferred to nitrocellulose paper, and probed with anti-β2GPI antibody. As a control marker, we used human β2GPI.
Association of Labeled β2GPI With Human U937 Myelomonocytic Cell Line and HUVECs
U937 (a monocyte/macrophage-like cell line) cells were grown in complete culture medium (RPMI supplemented with 10% fetal calf serum). Twenty-four hours before the experiments, the cells were washed with sterile PBS and transferred to serum-free medium. Cells (2×106) in RPMI with 1% BSA (in triplicates) were incubated with 125I-β2GPI (40 000 cpm) alone or in the presence of nonlabeled β2GPI, nLDL, or copper-oxLDL (all 200 μg/mL) for 3 hours at 37°C in a 5% CO2 incubator. After extensive washings with PBS, cells were lysed with 2.5N NaOH, and residual radioactivity was determined by a γ-counter.
Human umbilical vein endothelial cells (HUVECs) were isolated from normal-term umbilical cord veins by collagenase perfusion and cultured in 6-well plates under standard conditions.8 Twenty-four hours before the experiments, cells were washed and incubated in serum-free RPMI. The cells were detached by mechanical scrapping and resuspended in RPMI 1640 supplemented with 1% BSA. The assay for β2GPI association was continued as described for the U937 cells.
Calculations were done by ANOVA.
We found β2GPI to be expressed in all slides containing atherosclerotic lesions. Immunoperoxidase staining revealed the presence of β2GPI scattered in most areas of the lesions; however, it was most abundantly observed in the subendothelial regions and in the intimal-medial junction (Figure 1⇓, A and B). Staining was observed with both the polyclonal and monoclonal anti-β2GPI antibodies, which recognize distinct epitopes in native structure of human β2GPI. No staining was evident with control primary or secondary antibodies (Figure 1C⇓).
We also found that CD4-positive lymphocytes densely infiltrated the areas in which β2GPI was expressed, whereas a relative paucity of CD8-positive cells was evident. When we applied antibodies to adhesion molecules (intercellular adhesion molecule-1 [ICAM-1], vascular cell adhesion molecule-1 [VCAM-1], and E-selectin), we found that they were more equally distributed throughout the lesions and did not colocalize with β2GPI.
To confirm the presence of β2GPI in the lesions, taking into account the shortcomings of the immunohistochemical study, we used Western blot. Special care was taken to dispose of potential contamination of the lesions with serum β2GPI by repeated washings before the extraction. We confirmed the presence of β2GPI by observing a clear band in the 50-kDa region (data not shown).
Because endothelial cells and monocyte-derived macrophages are among the principal cells of atherosclerotic lesions, we wished to investigate the possibility that these cells were involved in β2GPI “consumption”/association. We found that 125I-β2GPI was rapidly taken up by U937 cells. When incubated with oxLDL (200 μg/mL), binding of 125I-β2GPI was inhibited by ≈56% (P<0.0001), whereas only 17% inhibition was observed with nLDL (P=NS) (Figure 2A⇓). Similar results were obtained with HUVECs, namely, β2GPI was significantly inhibited by oxLDL (53.4%; P<0.0001) and to a lesser extent by nLDL (13.6%; P=NS) (Figure 2B⇓).
The aim of the present study was to determine the presence and localization of β2GPI in human atherosclerotic lesions. We used 2 independent methods for immunodetection and lent support to our observations by exploring an association of β2GPI with cellular constituents of human lesions in vitro.
We have shown that β2GPI is abundantly present in human atherosclerotic plaques from carotid arteries. Although randomly expressed in the different layers of the plaque, it was found to be most prominent in subendothelial regions and in the intimal-medial border of the lesions. β2GPI colocalized with CD4-positive lymphocytes, whereas CD8-positive cells were rare. Western blotting confirmed the presence of β2GPI within the atherosclerotic lesions.
When U937 cells and HUVECs were assayed for their association with β2GPI, we found that both cell types, which represent principal cellular components of human lesions, were able to incorporate the radioiodinated protein. Interestingly, copper-oxLDL but not nLDL was highly specific in inhibiting the uptake of labeled β2GPI in both cell sources. Because β2GPI is a normal protein component that exists in the circulation, we washed the tissue specimens extensively and allowed the cultured cells to remain in serum-free conditions (devoid of β2GPI) for ≥24 hours before performance of the assays.
The significance of the detection of human β2GPI within human plaques can be determined with certainty only after a more elaborate understanding of the functional role of this protein has been achieved. There is evidence that β2GPI, which is principally synthesized by the liver, may act as a natural anticoagulant and possess antiatherogenic properties.9 This notion is indirectly supported by the observation that generation of an immunologically induced anti-β2GPI response (antagonizing the plasma protein) is associated with enhanced early atherosclerosis.5
However, when circulating autoreactive anti-β2GPI cells or anti-β2GPI antibodies are present and encounter their target antigen within preexisting lesions, they may contribute to acceleration of the ongoing local inflammatory reaction. The establishment of an immune response to β2GPI could be understood in view of the observation that the protein can be taken up and perhaps processed in a major histocompatibility complex–dictated manner by antigen-presenting cells (eg, endothelial cells and macrophages). By analogy, Stemme et al10 were recently able to obtain T-cell lines from human atherosclerotic plaques that recognize oxLDL. Such T cells are likely to derive from the circulation and are probably capable of secreting proinflammatory cytokines that may influence atherogenesis. These findings are supported by studies that show a requisite role of β2GPI in the activation of endothelial cells and platelets by aPL,11 12 13 and thus reinforce its role as a target of immune-mediated attack when presented within atherosclerotic lesions.
In conclusion, β2GPI is abundantly expressed in human atherosclerotic lesions and can be incorporated by endothelial cells and monocyte/macrophages. The results of the present study demonstrate that further investigation of the potential implications of local β2GPI accumulation on the progression of atherosclerosis is needed.
Reprint requests to Yehuda Shoenfeld, Department of Medicine B, Sheba Medical Center, Tel-Hashomer 52621, Israel.
Drs George and Harats contributed equally to this article.
- Received October 30, 1998.
- Revision received March 5, 1999.
- Accepted March 8, 1999.
- Copyright © 1999 by American Heart Association
George J, Afek A, Gilburd B, Aron-Maor A, Shaish A, Levkovitz H, Blank M, Harats D, Shoenfeld Y. Induction of early atherosclerosis in LDL receptor deficient mice immunized with β2 glycoprotein I. Circulation. 1998;15:1108–1115.
Gharavi AE, Sammaritano LR, Wen J, Elkon KB. Induction of antiphospholipid antibodies by immunization with β2 glycoprotein I (apolipoprotein H). J Clin Invest. 1992;90:1105–1109.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. J Clin Invest. 1973;52:2745–2756.
Stemme S, Faber B, Holm J. T lymphocytes from human atherosclerotic plaques recognize oxidized low density lipoprotein. Proc Natl Acad Sci U S A. 1995;92:3893–3897.
Simantov R, LaSala JM, Lo SK, Gharavi AE, Sammaritano LR, Salmon JE, Silverstein RL. Activation of cultured endothelial cells by antiphospholipid antibodies. J Clin Invest. 1995;96:2211–2219.
George J, Blank M, Levy Y, Meroni PL, Damianovich M, Tincani A, Shoenfeld Y. Differential effects of anti-β2GPI antibodies on endothelial cells and on the manifestations of experimental antiphospholipid syndrome. Circulation. 1998;97:900–906.
Shi W, Chong BH, Chesterman CN. β2-glycoprotein-I is a requirement for anticardiolipin antibodies binding to activated platelets: differences with lupus anticoagulants. Blood. 1993;81:1255–1261.β2-Glycoprotein I (β2GPI) is a natural in vitro anticoagulant present in the plasma. It has been recognized recently as an antigenic target of some autoimmune antiphospholipid antibodies. The immune response toward β2GPI has been shown to be associated with enhanced early atherosclerosis. In the present article, we demonstrate by immunohistochemistry and immunoblotting that β2GPI is present in atherosclerotic plaques and colocalizes with CD4-positive lymphocytes. In vitro, both endothelial and monocytic cells bind labeled-β2GPI, and the process is inhibited by oxidized LDL and not native LDL. Thus, β2GPI can serve as a target for an immune-mediated response in the vicinity of atherosclerotic lesions, thereby promoting atherogenesis.