Adoptive Transfer of β2-Glycoprotein I–Reactive Lymphocytes Enhances Early Atherosclerosis in LDL Receptor–Deficient Mice
Background—It has been proposed that autoimmune factors can influence the progression of atherosclerosis. We have previously shown that immunization of LDL receptor–deficient (LDL-RD mice) with β2-glycoprotein I (β2GPI; a principal target of “autoimmune” antiphospholipid antibodies) enhances early atherosclerosis. In the present study, we tested the hypothesis that adoptive transfer of β2GPI-reactive T cells can accelerate fatty streak formation in LDL-RD mice.
Methods and Results—LDL-RD mice were immunized with human β2GPI. An additional group of mice were immunized with β2GPI and boosted with the same antigen 3 weeks later. Control mice with immunized with human serum albumin. Lymphocytes obtained from the draining lymph node cells or from splenocytes of β2GPI- or human serum albumin–immunized mice were stimulated in vitro with β2GPI or with the mitogen concavalin A, respectively. The cultured lymphocytes were transferred intraperitoneally to syngenic LDL-RD mice, and the mice were fed a high-fat “Western” diet for 5 weeks until death. Mice injected with lymphocytes from draining lymph nodes or spleens of β2GPI-immunized animals displayed larger fatty streaks than those induced by control treated animals. T-cell–depleted splenocytes from β2GPI were unable to promote lesion formation in the mice.
Conclusions—The present study provides the first direct evidence for a role of antigen (β2GPI)-reactive T cells in the promotion of fatty streaks in mice.
Atherosclerosis is a multifactorial process that is the result of an interaction between different cellular elements, including lymphocytes, macrophages, endothelial cells, and smooth muscle cells.1 2 Traditional risk factors (eg, dyslipidemia, diabetes, hypertension, and smoking) have been solely thought to underlie the progression of atherosclerosis. However, the data that arise from clinical studies with an aim to ameliorate these risk factors have shown that they were not sufficient to completely suppress atherosclerosis. Thus, interest has focused on additional factors that could influence atherogenesis and possibly be a target for interventional manipulations.
The involvement of the immune system in atherosclerosis has been studied more extensively in recent years.3 4 The assumption that the immune system could influence atherosclerosis is based on findings of immune-potent cells (activated lymphocytes, antigen presenting cells) within atherosclerotic plaques and from the effect of manipulation of the immune system on atherogenesis in animals regardless of the effect of hyperlipidemia.5 6 An even more recent hypothesis holds that “footprints” of autoimmunity can be detected within the atherosclerotic lesion (summarized in Wick et al4 ). Accordingly, heat shock protein (HSP) 60/65 or modified lipoproteins could be targets of an autoimmune-mediated attack that can influence the progression of the atherosclerosis. The mechanisms that lie at the base of these apparently autoimmune processes include either antigenic mimicry between microbial HSP65 and host HSP60 or “altered self,” in which self-phospholipids (eg, oxidized LDL [oxLDL]) undergo oxidative modification, rendering them immunogenic.7 However, the multifactorial nature of the atherosclerotic lesion dictates the consideration of a larger number of involved autoantigens.
Patients with systemic lupus erythematosus develop premature atherosclerosis that is not exclusively accounted for by an altered lipid profile and immunosuppressive therapy.8 9 Approximately half of patients with systemic lupus erythematosus possess antiphospholipid antibodies, thus classifying them into a subgroup of patients with a hypercoagulable disorder: the secondary antiphospholipid syndrome.10 β2-Glycoprotein I (β2GPI) is a positively charged circulating plasma protein that has been suggested to be a principal autoantigen in patients with antiphospholipid syndrome.11 In a set of recent studies, we showed that β2GPI is present in human atherosclerotic plaques12 and that immunization of transgenic atherosclerosis-prone mice with the respective protein enhances fatty streak formation.13 14 However, despite the reproducibility of this model that encompasses immune-mediated involvement in atherogenesis, no mechanistic explanation could be provided regarding the acceleration of the lesions.
In the present report, we provide evidence to support the contribution of antigen (β2GPI)-specific cellular immunity to the progression of fatty streak formation in LDL receptor–deficient (LDL-RD) mice.
LDL-RD mice (hybrids of a cross between the C57BL/6J and 129Sv strains) were previously created with homologous recombination as described by Ishibashi et al.15 All mice that were used in the experiment were females of age 6 weeks. The mice were obtained from Jackson Laboratories and bred at the local animal facility.
The mice were fed either a normal chow diet that contained 4.5% fat by weight (0.02% cholesterol) or an atherogenic diet (Western-type diet, TD 96125; Harlan Teklad; 42% of calories from fat, 43% from carbohydrates, and 15% from protein). The mice were maintained on a 12-hour dark/light cycle and were allowed access to food and water ad libitum. Mice were maintained in a conventional environment.
Antigens and Antibodies
Human β2GPI was purified from the serum of a healthy adult as previously described by Gharavi et al.16 Rat monoclonal anti-mouse CD4+ and anti-mouse CD8α were obtained from PharMingen. MCA 497 (F4/80) anti-mouse macrophages were obtained from Serotec. Antibodies to CD3 and anti–interleukin (IL)-2 receptor and actin were from DAKO.
Polyclonal antibodies to mouse β2GPI were generated through repeated immunizations of rats with mouse β2GPI. The antibodies were primarily reactive with mouse β2GPI, with minimal cross-reactivity with human β2GPI.
Study Protocol and Proliferation Assays
In the first experiment, 6-week-old LDL-RD mice (n=10) were immunized subcutaneously in the hind footpad with a single dose of human β2GPI (50 μg/mouse) emulsified in incomplete Freund’s adjuvant (IFA). Control mice were immunized with human serum albumin (HSA; 50 μg in IFA/mouse). We previously observed (Gharavi et al16 ; unpublished observations) that immunization with ovalbumin and with HSA did not enhance atherosclerosis in LDL-RD mice compared with nonimmunized mice. Ten days after the initial immunization, the mice were killed. Their draining lymph node cells were collected (as described later) and incubated with β2GPI (10 μg/mL) in complete culture medium (RPMI 1640) on 24-well plates (Nunc). HSA-injected animals were treated in a similar manner. Three days later, the lymphocytes obtained from β2GPI- or HSA-immunized mice were collected from the plates, and 107 cells were transferred intraperitoneally into syngenic 6-week-old LDL-RD mice (n=8 per group).
In the second experiment, 6-week-old LDL-RD mice were immunized with 50 μg/mL β2GPI or HSA (emulsified in IFA) and immunity boosted 3 weeks later with a 25 μg/mL concentration of the primary antigen in IFA. Ten days after the boost, the spleens were removed and the splenocytes were incubated with concavalin A (2.5 μg/mL; Sigma Chemical Co). Three days later, the splenocytes from β2GPI- or HSA-immunized mice were collected, and 107 cells per mouse were transferred intraperitoneally into syngenic 6-week-old LDL-RD mice (n=9 per group).
As an additional control for the experiment, splenocytes from β2GPI- or HSA-immunized (and boosted) mice were depleted of their T cells with the use of nylon-wool columns. T-cell–depleted splenocytes were grown in culture for 3 days and transferred to syngenic 6-week-old LDL-RD mice (1×107 cells per mouse).
Before cell transfer, all preparations were washed 3 times with PBS.
Mice from all groups that received cells were fed a Western-type diet immediately after the transfer and killed 4 weeks later at the age of 10 weeks.
In a separate set of parallel studies, draining inguinal lymph node cells (experiment 1) or splenocytes (experiment 2) were collected from β2GPI- or HSA-immunized mice that had been killed 10 days after the primary immunization (experiment 1) or 10 days after boost (experiment 2), respectively. The assays were performed as previously described13 with minor modifications. Cells (2×105 well) were incubated in triplicate (on 96-well flat-bottomed plates) for 72 hours in 0.2 mL complete culture medium (RPMI 1640 plus 0.5% inactivated mouse sera) in the presence or absence of β2GPI. Proliferation was measured according to the incorporation of [3H]thymidine into DNA during the last 12 hours of incubation. The results were computed as stimulation index (SI): the ratio of the mean cpm of the antigen to the mean background cpm obtained in the absence of the antigen.
Preparation and Characterization of Cell Preparations
Total splenocytes were obtained from mice, and single-cell suspension was made. T cell–depleted preparations were made with nylon/wool columns (Eldan Technologies) according to the manufacturer’s instructions. The cells eluted from the columns were termed “T cell depleted.” To achieve maximal T-cell depletion, spleen cells were passed twice on the columns. T cells (detected by fluorescent activated cell sorter with anti-CD3 antibodies) were not detectable after 2 passages through the columns in the T cell–depleted preparations. Total splenocyte preparations contained ≈30% to 50% CD3-positive cells. Lymph node preparations contained >90% T cells.
Cholesterol Level Determinations
At the end of the experiment, 1 to 1.5 mL blood was obtained via cardiac puncture. Total plasma cholesterol levels were determined with an automated enzymatic technique (Boehringer Mannheim).
Detection of Anti-β2GPI Antibodies
Anti-β2GPI antibodies were detected by ELISA with β2GPI (10 μg/mL) for coating and performed as previously described.13
Interferon-γ, IL-4, and IL-4 Levels in the Culture Medium by ELISA
Levels of interferon-γ (IFN-γ), IL-4, and IL-10 were determined by an ELISA kit (PharMingen) according to the manufacturer’s instructions.
Assessment of Atherosclerosis
Quantification of atherosclerotic fatty streaks lesions was made through the calculation of the lesion size in the aortic sinus as previously described17 with a few modifications. Briefly, the heart and upper section of the aorta were removed from the animals, and the peripheral fat was cleansed carefully. The upper section was embedded in OCT medium and frozen. Every other section (10 μm thick) throughout the aortic sinus (400 μm) was taken for analysis. The distal portion of the aortic sinus is recognized by the 3 valve cusps that are the junctions of the aorta to the heart.
The extent of atherosclerosis was evaluated at the level of the aortic sinus. Processing and staining of the tissue with Oil-red O were carried out according to Paigen et al.17 Lesion area was quantified according to the method of Rubin et al.18
Immunohistochemistry of Atherosclerotic Lesions
Immunohistochemical staining for CD4, CD8, CD3, IL-2 receptor, and macrophages were performed on 5-μm-thick aortic sinus frozen sections. The sections were fixed for 4 minutes in methanol at −20°C, followed by a 10-minute incubation with ethanol at −20°C. The sections were then blocked with nonimmune goat serum for 15 minutes at room temperature, followed by incubation with CAS blocking reagent for 30 minutes at room temperature. Subsequently, the rat monoclonal anti-mouse antibodies were added for 1 hour at room temperature. After washings, affinity-purified biotinylated rabbit anti-rat IgG antibodies (Jackson Laboratories) were added for 30 minutes at room temperature. After washings, the slides were incubated with 0.3% H2O2, followed by additional rinses and incubation with streptavidin-peroxidase conjugate for 30 minutes at room temperature. After washings, the slides were developed with 3-amino-9-ethylcarbonasole substrate (DAKO) for 15 minutes. Sections were counterstained with hematoxylin. Spleen sections were used as a positive control. Staining in the absence of the first or second antibody was used as a negative control.
A 1-way ANOVA was used to compare independent values. P<0.05 was accepted as statistically significant. Results are presented as mean±SD.
Cholesterol Levels in LDL-RD Mice Are Not Influenced by Lymphocyte Transfer
Mean total cholesterol levels in mice transferred with β2GPI-reactive lymph node cells were 810±110 mg/dL compared with 833±125 mg/dL for control mice (P=NS) at the end of the study. Differences were also negligible between cholesterol levels in β2GPI-injected splenocytes (798±102 mg/dL) and HSA-injected mice (806±110 mg/dL). Cholesterol values did not differ significantly between the T-cell–depleted splenocytes from β2GPI- or HSA-immunized mice.
β2GPI Reactivity of Lymphocytes Used for the Transfer Experiments
The mean SI of lymphocytes obtained from draining lymph nodes of β2GPI-immunized mice primed with β2GPI (25 μg/mL) was 4.73±0.70 compared with 0.97±0.41 in control immunized mice (Figure 1A⇓). The mean proliferation index was also significant in splenocytes from mice that were repeatedly immunized with β2GPI (mean±SD 3.8±0.61) compared with controls (1.1±0.35) (Figure 1B⇓).
Cytokine Secretion of Lymph Node Cells
A study of the cytokine-secreting profile of lymph node cells from β2GPI-immunized mice disclosed a ≈5-fold enhancement of IFN-γ secretion into the cultured medium on in vitro stimulation with β2GPI (Figure 2A⇓). Although a similar trend was evident in control injected mice (after in vitro stimulation with β2GPI), it was of a significantly lesser extent. However, although nonstimulated IFN-γ levels were reduced in the β2GPI-immunized group, final IFN-γ levels in the presence of β2GPI did not differ between study and control groups. The reduced IFN-γ levels in the nonstimulated β2GPI-immunized group may have resulted from a lack of sufficient drive (sufficient β2GPI amounts) to these cells that are more β2GPI “educated” (due to prior β2GPI immunization).
IL-4 secretion was reduced to a similar degree on (β2GPI) stimulation of lymph node cells from β2GPI- and HSA-immunized mice (Figure 2B⇑).
IL-10 production was increased after the stimulation of β2GPI-reactive and control lymphocytes with β2GPI (Figure 2C⇑).
Fatty Streaks Are Enhanced in LDL-RD Mice Transferred With β2GPI-Reactive Lymphocytes
Fatty streak formation was enhanced in the mice injected with β2GPI-primed lymph node cells (mean lesion size 68 130±17 730 μm2) compared with mice transferred with lymphocytes obtained from HSA-immunized mice (20 100± 4410 μm2; P<0.05) (Figure 3A⇓). Fatty streaks were also accelerated in mice transferred with total splenocytes from β2GPI-primed mice (69 750±9780 μm2) compared with mice injected with control splenocytes (20 000±3876 μm2; P<0.001) (Figure 3B⇓).
In contrast, no statistically significant differences were evident between the extent of aortic sinus fatty streaks in mice transferred with T-cell–depleted splenocytes from β2GPI-immunized mice (19 100±5646 μm2) compared with mice injected with T-cell–depleted splenocytes from control immunized mice (15 500±1458 μm2; P=0.46) (Figure 3C⇑).
The composition of the early lesions did not differ between the mouse groups that we studied. The total macrophage content was in the range of 50% to 60% per lesion. T-lymphocytes were scarce within the lesions from all groups, representing 0% to 5% of the total plaque area; most were CD3+. IL-2 receptor–expressing cells were not evident in the lesions.
Anti-β2GPI Antibody Levels in the Recipient Mice
Antibodies to β2GPI were not detectable in the sera of any recipient mice at death.
β2GPI Is Abundantly Present in Fatty Streaks of LDL-RD Mice
By using a polyclonal rat anti-mouse β2GPI antibody and a monoclonal anti-human anti-β2GPI that cross-reacts with the murine protein, we observed that β2GPI is abundantly present in fatty streaks of all LDL-RD mice regardless of the manipulation used (Figure 4⇓). The presence of β2GPI was detected in the plaques from the early fatty streaks until the progression to the more complicated lesions. β2GPI was found intracellularly in cells that corresponded to smooth muscle cells, macrophages, and endothelial cells (data not shown). However, the protein was also abundant extracellularly and on the luminal surfaces of endothelial cells, predominantly at the sites of lesion formation.
Despite the accumulating data on the role of autoimmune factors in atherogenesis, the evidence remains circumstantial. (1) HSP60, oxLDL, and β2GPI have been demonstrated within atherosclerotic lesions of animals and humans (see summaries4 7 12 ). (2) IgG anti-HSP65 and oxLDL autoantibodies have been shown to be associated with atherosclerosis in humans.19 20 (3) Immunization of animals with oxLDL was shown to reduce, whereas injection of HSP65 and β2GPI was shown to enhance, atherosclerosis.13 14 21 22 23 These data do not provide sufficient insight into the effector cells and mechanisms that mediate the progression of atherogenesis rather than implying that the cellular and humoral arms of the immune system could be involved.
In the present study, we provide the first direct evidence that antigen-specific cellular immunity can influence the progression of the fatty streak in mice. Previously, we showed that the immunization of LDL-RD mice with β2GPI increased early fatty streak formation in chow diet–fed fat animals and marginally influenced lesions induced with a cholate-containing high-fat diet.13 However, the enhancing effect was not investigated. Here, we obtained β2GPI-reactive lymphocytes from the draining lymph nodes and from spleens of immunized animals and demonstrated their ability to enhance atherogenesis. It is important to mention that cholesterol levels did not differ between groups, suggesting that the accelerating effect on atherogenesis was not mediated by elevated lipid levels.
We observed that lymphocytes obtained from draining lymph nodes, which are mostly T cells, were able to increase fatty streak formation, as were β2GPI-primed splenocytes (which also contain T cells). However, when splenocytes were depleted of their T cells, fatty streaks in mice to which cells were administered did not differ between β2GPI-immunized and control groups. These observations support a primary role for T cells in mediation of the proatherogenic effect.
Next, we hypothesized that β2GPI T cells can influence atherosclerosis by reaching the lesion, which should contain the respective antigen. We have previously shown that β2GPI was present in human plaques and incorporated by endothelial cells and macrophages.12 Here, we demonstrated with immunohistochemistry that murine β2GPI was abundantly present both intracellularly and extracellularly at the early atherosclerotic lesion (colocalizing with macrophages, endothelial cells, and smooth muscle cells). Thus, the signal for recruitment of β2GPI-reactive cells was present in murine lesions and could subsequently serve to allow for ligation of the T-cell receptor in the transferred cells. Because β2GPI is a highly conserved molecule with an ≈80% degree of sequence homology to human β2GPI, the murine and human proteins are both acceptable antigens for the in vitro assays.
If antigen-specific T cells are capable of influencing fatty streak formation, they may do so by altering their cytokine-secretion pattern. A paradigm of functional dichotomy of T cells into T-helper 1 (Th1) and T-helper 2 (Th2) cells was extensively developed in the past decade. Accordingly, T cells can be subgrouped according to their cytokine-secreting profile into Th1 cells (secreting principally proinflammatory cytokines such as IFN-γ and tumor necrosis factor-α) and Th2 (secreting anti-inflammatory cytokines such as IL-4 and IL-10).24
In experimental atherosclerosis, apoE-deficient mice have been shown to display a transition from a Th1 toward a Th2 pattern.25 In general, it appears that immune reactions that favor a Th1 pattern lead to enhanced atherogenesis. Accordingly, IFN-γ receptor–deficient mice crossed with the apoE-deficient mice yield a double knockout mouse model with a 50% reduction in atherosclerosis.26 In addition, the treatment of apoE mice with IL-12 leads to increased IFN-γ secretion (a switch to Th1) and accelerated atherosclerosis.27 A very recent study by Mallat et al28 reinforces these observations, showing that IL-10 has a protective role in murine atherosclerosis. In the present report, we examined the cytokine-secretion pattern of the transferred lymphocytes on contact with their respective autoantigen. Our observations do not support straightforward evidence for the role of Th1/Th2 shifts in the acceleration of atherosclerosis, because final β2GPI-stimulated levels of IFN-γ, IL-4, and IL-10 did not differ between the β2GPI and control immunized mice.
Evidence for the involvement of T cells in atherogenesis in animals and humans derives from several lines of studies. (1) Activated T cells have been demonstrated in early and late atherosclerotic lesions of animals and humans.29 30 In addition, T cell lines from plaques have been grown and shown to be reactive with modified LDL.31 (2) Global T-cell suppression with cytotoxic agents has been shown to influence the progression of atherosclerosis.5 6 (3) Mice that are deficient in both recombinase activating gene (required for T- and B-cell maturation) and apoE exhibited reduced early, but not late, atherosclerotic lesion formation.32 The most significant observation in the present study is that antigen-specific T cells have a role in fatty streak formation. Thus, cellular autoimmunity may be a clinically significant factor that can play a role in the pathogenesis of early atherosclerosis.
↵1 These two authors contributed equally to this work.
- Received March 3, 2000.
- Revision received May 15, 2000.
- Accepted May 16, 2000.
- Copyright © 2000 by American Heart Association
Wick G, Schett G, Amberger A, et al. Is atherosclerosis an immunologically mediated disease? Immunol Today. 1995;16:27–33.
Roselaar SE, Schonfeld G, Daugherty A. Enhanced development of atherosclerosis in cholesterol fed rabbits by suppression of cell mediated immunity. J Clin Invest. 1995;96:1389–1394.
Witztum J, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest. 1991;88:1785–1792.
Kandia DA, Krilis SA. Beta2-glycoprotein I. Lupus. 1994;3:207–212.
George J, Harats D, Gilburd B, et al. Immunolocalization of β2-glycoprotein I (apolipoprotein H) to human atherosclerotic plaques: potential implications for lesion progression. Circulation. 1999;99:2227–2230.
George J, Afek A, Gilburd B, et al. Induction of early atherosclerosis in LDL-receptor–deficient mice immunized with β2-glycoprotein I. Circulation. 1998;15:1108–1115.
Ishibashi S, Brown MS, Goldstein JL, et al. Hypercholesterolemia in LDL receptor knockout mice and its reversal by adenovirus mediated gene delivery. J Clin Invest. 1993;92:883–893.
Gharavi AE, Sammaritano LR, Wen J, et al. Induction of antiphospholipid antibodies by immunization with β2 glycoprotein I (apolipoprotein H). J Clin Invest. 1992;90:1105–1109.
Palinski W, Miller E, Witztum JL. Immunization of low density lipoprotein (LDL) receptor-deficient rabbits with homologous malondialdehyde-modified LDL reduces atherogenesis. Proc Natl Acad Sci U S A. 1995;92:821–825.
Xu Q, Dietrich H, Steiner HJ, et al. Induction of arteriosclerosis in normocholesterolemic mice rabbits by immunization with heat shock protein 65. Arterioscler Thromb. 1992;12:789–799.
Zhou X, Paulsson G, Stemme S, et al. Hypercholesterolemia is associated with a T helper (Th) 1/Th2 switch of the autoimmune response in atherosclerotic apo E-knockout mice. J Clin Invest. 1998;101:1117–1225.
Lee TS, Yen HC, Pan CC, et al. The role of interleukin 12 in the development of atherosclerosis in apoE-deficient mice. Arterioscler Thromb Vasc Biol. 1999;19:734–742.
Mallat Z, Besnard S, Duriez M, et al. Protective role of interleukin-10 in atherosclerosis. Circ Res. 1999;85:17–24.
Roselaar SE, Kakkanathu PX, Daugherty A. Lymphocyte populations in atherosclerotic lesions of apo-E −/− and LDL receptor −/− mice: decreasing density with disease progression. Arterioscler Thromb Vasc Biol. 1996;16:1013–1018.
Stemme S, Faber B, Holm J, et al. T lymphocytes from human atherosclerotic plaques recognize oxidized low density lipoprotein. Proc Natl Acad Sci U S A. 1995;92:3893–3897.
Dansky HM, Charlton SA, Harper MM, et al. T and B lymphocytes play a minor role in atherosclerotic plaque formation in the apolipoprotein E-deficient mouse. Proc Natl Acad Sci U S A. 1997;94:4642–4646.