Donate Help Contact The AHA Sign In Home
American Heart Association
Arteriosclerosis, Thrombosis, and Vascular Biology
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Arteriosclerosis, Thrombosis, and Vascular Biology. 1998;18:1972-1982

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Freigang, S.
Right arrow Articles by Palinski, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Freigang, S.
Right arrow Articles by Palinski, W.
(Arteriosclerosis, Thrombosis, and Vascular Biology. 1998;18:1972-1982.)
© 1998 American Heart Association, Inc.

Original Contributions

Immunization of LDL Receptor–Deficient Mice With Homologous Malondialdehyde-Modified and Native LDL Reduces Progression of Atherosclerosis by Mechanisms Other Than Induction of High Titers of Antibodies to Oxidative Neoepitopes

Stefan Freigang; Sohvi Hörkkö; Elizabeth Miller; Joseph L. Witztum; Wulf Palinski

From the Department of Medicine, University of California San Diego, La Jolla, Calif.

Correspondence to Wulf Palinski, MD, University of California San Diego, Department of Medicine, 0682, 9500 Gilman Dr, BSB 1080, La Jolla, CA 92093-0682. E-mail wpalinski{at}ucsd.edu


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Abstract—We and others previously showed that immunization of rabbits with different forms of oxidized low density lipoprotein (LDL) significantly reduced atherogenesis. We now investigated the effect of continued immunization on atherosclerosis in LDL receptor–deficient (LDLR-/-) mice to determine whether a similar reduction of atherosclerosis occurred in murine models and whether this was due to humoral immune responses, ie, formation of high titers of antibodies to oxidation-specific epitopes. Three groups of LDLR-/- mice were repeatedly immunized with homologous malondialdehyde-modified LDL (MDA-LDL), native LDL, or phosphate-buffered saline (PBS) for 7 weeks. Extensive hypercholesterolemia and accelerated atherogenesis were then induced by feeding a cholesterol-rich diet for 17 weeks, during which immunizations were continued. Binding of immunoglobulin (Ig) M and IgG antibodies, as well as IgG1 and IgG2a isotypes, to several epitopes of oxidized LDL were followed throughout the study. After 24 weeks of intervention, atherosclerosis in the aortic origin was significantly reduced by 46.3% and 36.9% in mice immunized with MDA-LDL and native LDL, respectively, compared with PBS (133 558 and 157 141 versus 248 867 µm2 per section, respectively). However, the humoral immune response to oxidative neoepitopes in the MDA-LDL group was very different from that of the LDL or PBS group. IgG antibody binding to MDA-LDL and other epitopes of oxidized LDL, such as oxidized phospholipid (cardiolipin), oxidized cholesterol, or oxidized cholesteryl linoleate, but not native LDL, increased markedly in mice immunized with MDA-LDL, but not in mice immunized with native LDL or PBS. In the MDA-LDL group, both T helper cell (Th)2–dependent IgG1 antibody and Th1-dependent IgG2a antibody binding to oxidative neoepitopes increased significantly over time. The fact that mice immunized with both MDA-LDL and native LDL had a significant reduction in atherosclerosis, whereas only the MDA-LDL group developed very high titers of antibodies to oxidation-specific epitopes, suggests that the antiatherogenic effect of immunization is not primarily dependent on very high titers of antibodies to oxidation-specific epitopes but is more likely to result from the activation of cellular immune responses.


Key Words: arteriosclerosis • oxidized lipoproteins • immune system • autoantibodies • T cells


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Oxidation of lipoproteins and oxidative processes in general play an important role in the initiation and progression of atherosclerotic lesions. Oxidation of LDL occurs in a wide range of atherosclerotic lesions in humans and animal models,1 2 3 4 5 and by-products generated during this process, such as lipid peroxidation products and modified apoproteins, possess numerous properties that may contribute to atherogenesis (reviewed in References 6 through 86 7 8 ). These include cytotoxic damage to vascular cells, inhibition of vasodilation in response to nitric oxide, chemotactic recruitment of monocytes into the artery wall, phenotypic transformation of monocytes into macrophages, as well as the rapid uptake of oxidized LDL (OxLDL) by macrophages via scavenger receptors. More recently, it has also been recognized that oxidative processes may affect the regulation of gene expression by vascular cells and promote expression of a number of adhesion molecules and cytokines that may in turn contribute to lesion formation.8

The oxidation of LDL results in many structural modifications and the formation of a large number of neoepitopes.1 9 For example, reactive aldehydes generated during lipid peroxidation form covalent adducts with lysine and histidine residues of apoB or other closely associated proteins.6 7 These modified lysines are highly immunogenic, particularly when present on LDL.10 11 The immunogenicity of modified LDL has greatly facilitated generation of antibodies against "oxidation-specific epitopes," such as malondialdehyde (MDA)-lysines and 4-hydroxynonenal (4-HNE)–lysine.12 By analogy, OxLDL present in atherosclerotic lesions triggers a humoral immune response in vivo, and autoantibodies binding to various epitopes of OxLDL have been described in humans, rabbits, and mice.1 13 14 15 16 Extensive data have since accumulated suggesting that the titers of such antibodies may be of diagnostic and/or prognostic value (reviewed in Reference 1717 ). In a Finnish population in whom carotid atherosclerosis was followed by ultrasound over a 2-year period, the titer of autoantibody binding to MDA-LDL was significantly greater in subjects with rapid progression of atherosclerosis than in control subjects with minimal progression.18 Furthermore, the antibody titer was an independent predictor of progression of the disease. Most but not all subsequent epidemiological studies in humans also suggested that the titer of autoantibodies to MDA-LDL or copper-oxidized LDL (Cu-LDL) may be an indicator of the severity or rate of progression of the disease. For example, increased autoantibody titers were reported in patients with carotid atherosclerosis, coronary artery disease, diabetes, peripheral vascular disease, hypertension, and preeclampsia.19 20 21 22 23 24 25 26 27 Finally, a prospective study carried out in a more genetically homogeneous population, ie, LDL receptor–deficient (LDLR-/-) mice, demonstrated that autoantibody titers were directly correlated with the extent of aortic atherosclerosis.28

These observations raised the question whether humoral and/or cellular responses to OxLDL might not only reflect but also actively modulate the atherogenic process. The fact that lesions contain large numbers of immune-competent cells (macrophages and T cells), markers of their activation (interleukin-2 receptors, major histocompatibility complex [MHC] class II molecules), immunoglobulins, and terminal C5b-9 complement complexes has long suggested involvement of the immune system in atherogenesis (for a review, see Reference 2929 ). However, it was generally assumed that immune reactions would enhance lesion formation, as is clearly the case in transplant atherosclerosis.30 31 We and others have recently provided evidence that under particular circumstances, the opposite may be true. A study from our laboratory demonstrated that continuous immunization of LDLR-/- Watanabe heritable hyperlipidemic (WHHL) rabbits with MDA-modified homologous LDL resulted in very high titers of antibodies with specificity similar to that of naturally occurring autoantibodies and that this intervention significantly reduced the progression of atherosclerosis.32 Similar results were subsequently reported for cholesterol-fed New Zealand White rabbits33 and balloon-catheterized rabbits34 immunized with Cu-LDL. However, the mechanisms responsible for this beneficial effect remain unknown.

Crosses of atherosclerosis-susceptible murine strains, such as apoE-deficient and LDLR-/- mice, with murine strains with well-characterized immune deficiencies would offer excellent models in which to investigate whether the beneficial effects of immunization stem from the humoral and/or cellular system. In the current article, we demonstrate that immunization of LDLR-/- mice results in a reduction of atherogenesis and test the hypothesis that the beneficial effect is due to the formation of high-titered antibodies to oxidative neoepitopes.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Mice and Diets
Forty-four male LDLR-/- mice (C57BL/6Jx129Sv background),35 age 21 to 28 weeks, were selected from our breeding colony established from animals originally provided by The Jackson Laboratories, Bar Harbor, Me. Mice were divided into 3 groups matched for age, body weight, and initial plasma cholesterol levels. The first group (n=15) was immunized with homologous MDA-modified LDL, as described below (MDA-LDL group). The second group (n=15) was immunized with homologous "native" LDL (LDL group). The third group (n=14) was immunized with PBS (control group). Mice were initially fed nonatherogenic rodent chow (Harlan Teklad W860) for 7 weeks. Under these conditions, LDLR-/- mice are known to develop only moderate hypercholesterolemia and limited atherosclerosis.28 During this period, mice were subjected to the primary and 3 biweekly booster immunizations. For the remaining 17 weeks of the intervention period, mice were fed an atherogenic diet free of cholate. This diet contained 21.2% milkfat and initially (weeks 8 through 10) 1% cholesterol and later (weeks 11 through 24) 1.25% cholesterol (TD96121, Harlan Teklad). Mice were kept on a 12-hour day/night cycle and had unrestricted access to water and food. Three weeks before the start of the experiment and {approx}1 week after each booster immunization, 100 to 150 µL blood was obtained in heparinized tubes from the retro-orbital plexus of anesthetized (diethyl ether inhalation) mice. A plasma aliquot of these samples was used for preliminary measurements of antibody titers and for determination of plasma cholesterol and triglyceride levels. The remainder of the plasma was stored at -70°C for the final determination of all antibody titers under identical assay conditions. Plasma cholesterol and triglyceride levels were determined using an automated enzymatic assay (Boehringer Mannheim Diagnostics). The experimental protocol was approved by the Animal Subjects Committee of UCSD.

Preparation of Antigens
Murine LDL used as the immunogen (either in its native form or after modification with MDA) was freshly prepared for each immunization. A typical preparation (3.5±0.9 mg apoprotein) was obtained from 24 healthy LDLR-/- mice that had been fed a diet containing 1% cholesterol (Harlan Teklad TD95286) for 4 to 5 weeks. Blood was obtained from the vena cava of anesthetized (diethyl ether inhalation) mice. LDL (1.023<d<1.058 g/mL) was isolated by sequential ultracentrifugation in the presence of antioxidants and antiproteolytic agents.12 In brief, after the density was adjusted to 1.022 mg/mL with NaBr, plasma was centrifuged at 65 000 rpm for 4 hours at 10°C using a Beckman L7-65 ultracentrifuge with a 65.2 rotor. The supernatant was pipetted off, the density of the infranatant was adjusted to 1.063 with NaBr, and the LDL fraction was isolated by centrifugation at 44 000 rpm for 22 hours using a 50.3 rotor. LDL was then extensively dialyzed against PBS containing 2 mmol/L EDTA and sterile filtered.

An aliquot of each LDL preparation was modified with MDA, as previously described.12 The degree of modification of the lysine residues of apoB was determined using the trinitrobenzenesulfonic acid assay36 and ranged from 72% to 82% (mean, 79%). The extent of modification was also verified by comparing the electrophoretic mobility of MDA-LDL to that of native LDL on agarose gels.37 Native and MDA-LDLs were stored at 4°C and used within 2 weeks.

To assess antibody titers in murine plasma throughout the intervention period, human LDL was isolated from a pool of healthy donors12 and modified with MDA, Cu2+, or 4-HNE. Human MDA-LDL was generated in the same way as murine MDA-LDL. Cu-LDL was generated by incubating 100 µg LDL per mL PBS, pH 7.35, with 5 µmol/L CuSO4 at 37°C for 2 to 16 hours,12 dialyzing it against PBS containing EDTA, and concentrating it with an Amicon Centriflo membrane cone-type CF25. For the screening of antibodies, a mixture of LDLs oxidized with Cu2+ for 2 hours or 16 hours was used to ensure that both early- and late-oxidation epitopes would be present. 4-HNE–LDL was generated under reducing conditions, as previously described.12

Immunization
The primary immunization consisted of 50 µg of homologous, native LDL or MDA-LDL in 125 µL PBS suspended in an equal volume of Freund's complete adjuvant. Mice were inhalation-anesthetized, and the immunogen was injected subcutaneously into both inguinal areas. Booster immunizations consisted of 25 µg antigen in Freund's incomplete adjuvant injected intraperitoneally 2, 4, 6, and 8 weeks after the primary immunization and subsequently at 4-week intervals for 3 more months. Control animals were immunized with an equal volume of sterile PBS in Freund's complete or incomplete adjuvant to match immunizations with MDA-LDL and native LDL. Parallel experiments not included in this report showed that the amount of antigen and the frequency of booster immunizations were greater than necessary to induce high-titered antibodies in mice, but boosting was continued as described to maintain an experimental protocol similar to our previous study in rabbits.32

Determination of Antibody Binding to Native and Oxidized Lipoproteins
Titers of autoantibodies in murine sera binding to LDL, MDA-LDL, Cu-LDL, and 4-HNE–LDL were determined using a chemiluminescence enzyme immunoassay (CLEIA).16 38 In this assay, 5 µg/mL of the antigen in 50 mmol/L Tris-buffered saline (TBS), pH 7.5, containing 0.27 mmol/L EDTA, 0.02% NaN3, and 20 µmol/L BHT (dilution buffer) was added to each well of a 96-well, white, round-bottomed MicroFluor microtitration plate (Dynex Technologies) and incubated overnight at 4°C. Plates were washed 4 times with washing buffer (TBS containing 0.27 mmol/L EDTA, 20 µmol/L BHT, 0.02% NaN3, and 0.001% aprotinin) in an automated plate washer. Murine sera were diluted 1:10 to 1:1000 (as indicated in individual experiments) in dilution buffer containing 2% or 3% BSA, and 50 µL was added to each well and incubated for 1 hour at room temperature. After 4 washes, plates were incubated with 50 µL/well of an alkaline phosphatase–labeled goat anti-mouse IgG ({gamma}-chain specific) or alkaline phosphatase-labeled goat anti-mouse IgM (µ-chain specific) (Sigma) for 1 hour at room temperature. These antibodies were diluted in 1% BSA/TBS according to the supplier's specifications. IgG1 isotype antibody binding to the plated antigen was detected using an alkaline phosphatase–labeled rat monoclonal antibody to mouse IgG1 (clone LO-MG1-2, Zymed). IgG2a antibodies were detected with an alkaline phosphatase–labeled monoclonal rat antibody to mouse IgG2a (clone R19-15, PharMingen). After the plates were washed, 25 µL of a 50% solution of Lumi-Phos 530 (Lumigen) was added to each well, and plates were incubated for 1 to 2 hours at room temperature in the dark. Luminescence was determined using a Lucy 1 luminometer and WINLCOM software (Anthos Labtec Instruments). Antibody binding was measured as relative light units (RLUs) in 100 ms. Triplicate determinations were performed for each plasma sample. Measurement of antibody binding to a given antigen was done in a single assay. A high and a low standard serum was included on each plate of a given assay to detect potential variations between microtitration plates. The intra-assay coefficient of variation for these assays was 6% to 10%.

Determination of Antibody Binding to Oxidized Phospholipids and Cholesterol
The CLEIA used to determine antibody binding to cardiolipin differed from the above with regard to the plating of the antigen. We recently demonstrated that the standard assays used to measure "anticardiolipin" antibodies actually measure antibodies to oxidized cardiolipin (OxCL).38 In standard assays, cardiolipin is added to microtiter wells in ethanol and is plated by overnight evaporation of the solvent. Plated cardiolipin exposed to air is rapidly oxidized within a few minutes.38 39 To standardize the extent of oxidation, bovine heart cardiolipin (Avanti Polar Lipids) was diluted to 20 µg/mL in 100% ethanol, and 25 µL was added to each well. Plates were dried under air at room temperature ({approx}30 minutes) and air exposure was then continued for another hour. Under these conditions the unsaturated fatty acids of cardiolipin undergo extensive oxidation, leading to formation of oxidative neoepitopes recognized by anticardiolipin antibodies,38 39 40 as well as by some induced monoclonal antibodies to OxLDL.38 Absolute ethanol was added to blank wells, and binding to these wells was subtracted to obtain specific binding. Murine sera diluted with 2% BSA/TBS were added to the antigen-coated wells after the plates were washed, and the antibody binding to the OxCL was determined as described above.

Antibody binding to unoxidized and oxidized cholesterol, cholesteryl linoleate, and cholesteryl oleate was determined as follows: Antigens (Sigma) were solubilized in isopropanol at 10 µg/mL and plated at 25 µL/well. To determine binding to the oxidized forms of these antigens, cholesterol, cholesteryl linoleate, and cholesteryl oleate were oxidized by plating and allowing the antigens to be exposed to air overnight at room temperature after evaporation of the solvent. Plasma was then added the following morning. In contrast, binding to the unoxidized forms was achieved by adding the antigens to the wells, evaporating the solvent under argon, and then immediately adding plasma to the wells. Detection of bound antibodies was performed as described above, using labeled second antibodies to murine IgG or IgM.

Determination of Immune Complexes on Circulating Murine LDL
To detect circulating immune complexes between antibodies potentially induced by immunization or physiologically present autoantibodies and oxidative neoepitopes on LDL, a "sandwich" CLEIA was used. A rabbit anti-mouse apoB antiserum (G-485-3; a kind gift from Dr Helen Hobbs, University of Texas Southwestern Medical Center, Dallas, Tex) was diluted to 10 µg/mL and plated overnight at 4°C (capture antibody). Wells were then incubated with a 1:50 dilution of murine plasma for 2 hours at room temperature. After thorough washing, immunoglobulins bound to the captured LDL were detected with alkaline phosphatase–labeled goat anti-mouse IgG or IgM antibodies, as described above.

Evaluation of Atherosclerosis
The heart and aortic tree were perfusion-fixed with formal sucrose, dissected, and prepared for morphometry, as previously described.15 The extent of atherosclerosis was determined in en face preparations of the entire aortic tree, as well as in cross sections through the aortic origin, by computer-assisted image analysis described in detail in Reference 4141 . The measurement of cross sections through the aortic origin was done to provide a measure of lesion thickening, because surface measurements of Sudan IV–stained aortas cannot differentiate between lesions of different stages. Lesion size in the aortic origin was measured in ten 8-µm-thick hematoxylin/eosin-stained step sections (every sixth serial section over a distance of 480 µm, beginning with the appearance of at least 2 valve leaflets). All measurements of aortic surface areas were performed by the same investigator. Measurements of lesion areas in the cross sections of the aortic origin were also performed by the same operator. Because of the less-automated detection of intimal/medial delimitation, selected sections from each group were double checked by a second investigator.

Statistical Analysis
Results were analyzed by ANOVA and Student's unpaired t test. Data shown are mean±SEM. Data on the extent of atherosclerosis were also analyzed by nonparametric tests.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Effect of Intervention on Body Weight and Plasma Lipid Levels
Mice were of fairly mature age (21 to 28 weeks) at the start of the experiment, and the matched baseline body weights in the 3 groups were at adult levels (30.8±0.1 g). None of the interventions had a significant effect on body weight, which showed only a mild increase after the animals were put on the high-fat, high-cholesterol diet. The final body weights were 36.0±1.7 g in the group immunized with MDA-LDL, 43.0±2.3 g in the group immunized with native LDL, and 34.9±1.5 g in the control group immunized with PBS.

Plasma cholesterol levels were not affected by the primary immunization or the first 2 boosts and remained at the preimmunization level in all 3 groups (Figure 1Down). After the start of the high-cholesterol diet at week 7, plasma cholesterol levels increased in parallel to {approx}1400 mg/dL. Levels in the control group then remained relatively constant, whereas in the MDA-LDL and LDL groups a slight decrease was observed. However, the difference between the 3 groups did not reach statistical significance, except at the final time point. The time-averaged cholesterol levels (area under the cholesterol-versus-time curve divided by the entire length of the study) were also similar (928±77, 937±47, and 1084±70 mg/dL in the MDA-LDL, LDL, and PBS groups, respectively; NS). Triglyceride levels rose from 106±5 mg/dL at the start of the experiment to 608±140 mg/dL in the MDA-LDL group, 500±93 mg/dL in the LDL group, and 713±114 mg/dL in the PBS group (NS).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 1. Total plasma cholesterol levels in the 3 treatment groups throughout the intervention period. As described in Methods, mice were initially fed a low-cholesterol diet to allow for the development of significant antibody titers before the development of extensive atherosclerosis. Seven weeks after the primary immunization, mice were put on a high-fat, high-cholesterol diet (arrow). Blood was collected from each mouse at the time points indicated (usually 1 week after a booster immunization) and cholesterol levels determined as described in Methods. Each value represents mean±SEM of 15 mice in the MDA-LDL group, 13 mice in the LDL group, and 11 mice in the PBS group. Differences between the group immunized with MDA-LDL (squares) and the PBS group (triangles) were not significant, except at the final time point (P=0.05). Similarly, differences between the group immunized with LDL (circles) and the PBS group were only significant at 24 weeks (P=0.02).

Effect of Intervention on Atherosclerosis
After 24 weeks of intervention, the extent of atherosclerosis was determined by computer-assisted image analysis of 10 step sections through the aortic origin of each mouse. Statistical evaluation of data from all 3 groups suggested a normal distribution, and ANOVA indicated that the groups were different (P<0.01). Comparison of groups was therefore performed using Student's unpaired t test. As shown in Figure 2Down, immunization with MDA-LDL led to a significant reduction of lesion size compared with the control group (133 588±20 577 versus 248 867±38 207 µm2/section, P<0.01). Surprisingly, immunization with LDL also reduced the lesions. The mean lesion size in the LDL group was 157 141±61 680 µm2/section (P<0.05, compared with the control group). Comparisons of differences between intervention and control groups using nonparametric statistics also yielded significant differences between groups. Although the reduction in lesion size was less impressive in the LDL group than in the MDA-LDL group (36.9% versus 46.3%), the difference between these 2 groups was not statistically significant (P=0.49).



View larger version (35K):
[in this window]
[in a new window]
  		Figure 2. Extent of atherosclerosis in the aortic origin of LDLR-/- mice immunized with MDA-LDL, LDL, or PBS. The average size of atherosclerotic lesions per section in the aortic origin after 24 weeks of intervention was assessed by computer-assisted image analysis, as described in Methods. Values are mean±SEM. *P=0.009, **P=0.03, compared with the PBS group.

In our experience, exposure of LDLR-/- mice41 42 to plasma cholesterol levels >1000 mg/dL for only 3 to 4 months, as in the current experiment, only induces mild atherosclerosis that involves 7% to 10% of the aortic tree. We nevertheless also determined the percent of aortic surface area covered by Sudan IV–positive lesions in the 3 treatment groups. Consistent with previous data,41 42 <10% of the aortic surface was involved, and no significant differences in the percent of atherosclerotic surface area were found between the 3 experimental groups (MDA-LDL, 8.14±1.1%; LDL, 9.08±1.2%; and control, 8.4±1.3%). Although we did not have a control group that was not subjected to any immunization, it should be noted that the degree of atherosclerosis observed in the PBS group in either location was very similar to what we had previously seen in LDLR-/- mice on similar diets.28 41 42

As reported above, the differences in plasma cholesterol levels between the 3 groups were not statistically significant. Nevertheless, the absolute plasma cholesterol concentrations of the MDA-LDL and LDL groups were consistently lower than those of the control group (Figure 1Up), and their time-averaged cholesterol levels were also lower by {approx}150 mg/dL. To rule out the possibility that the reduction in atherosclerosis in the MDA-LDL and LDL groups might be accounted for by this difference in cholesterol exposure, the correlation between the time-averaged plasma cholesterol level (area under the curve divided by time; Figure 1Up) and the average size of atherosclerotic lesions in the aortic origin was determined by using the data of all mice from the 3 groups. As shown in Figure 3Down, a weak but significant correlation existed between the 2 parameters even at the very high cholesterol levels induced in this study (P<0.001). However, the slope of the regression line was shallow, and a difference in plasma cholesterol of 150 mg/dL could only have accounted for a small part (<35 000 µm2/section) of the reduction in atherosclerosis actually observed (-115 279 and -91 726 µm2/section in the MDA-LDL and LDL groups, respectively). Thus, immunization reduced atherogenesis by a distinct mechanism, in addition to a potential effect resulting from the slight lowering of plasma cholesterol.



View larger version (22K):
[in this window]
[in a new window]
  		Figure 3. Correlation between the time-averaged plasma cholesterol level and the average size of atherosclerotic lesions in the aortic origin. The time-averaged plasma cholesterol level (area under the curve [Figure 1Up] divided by time) and extent of atherosclerosis are shown for mice from the 3 treatment groups.

Humoral Immune Response to Immunization
Plasma samples from each mouse were obtained before immunization and throughout the study to follow the humoral immune response. Because homologous LDL (or MDA-LDL) was used to immunize the mice, antibodies against epitopes of native LDL should not be generated. Indeed, IgG antibody binding to native murine or human LDL was extremely low at any time point in all of the 3 experimental groups (data not shown).

The first model epitope of OxLDL tested was MDA-LDL. For these assays, we postulated that human MDA-LDL could be used as the antigen, because antibodies previously generated by immunizing mice and rabbits with homologous MDA-LDL recognized MDA-LDL prepared from LDL of different species.1 4 12 16 Preliminary binding assays indeed showed similar binding to mouse MDA-LDL (the actual immunogen) as to human MDA-LDL. As expected, immunization with MDA-LDL induced a >200 000-fold rise of specific IgG binding to MDA-LDL (FigureUp A) when plasma was tested at a 1:1000 dilution. In contrast, animals immunized with PBS showed only a 39% rise in absolute binding to MDA-LDL and animals immunized with native LDL, a 50% increase. After the initial rise, the binding to MDA-LDL remained fairly constant in the MDA-LDL group, despite additional booster immunizations, the switch to high-cholesterol diets, and the resulting atherogenesis.

We then determined the antibody binding to other oxidative neoepitopes. We began by determining binding to oxidized phospholipids by using OxCL as the antigen. We previously demonstrated that oxidized phospholipid epitopes are generated when LDL is oxidized.38 39 In the current study, the MDA-LDL group also demonstrated a dramatic increase in the binding of antibodies to OxCL, with a somewhat similar time course as the binding to MDA-LDL (Figure 4BDown).



View larger version (27K):
[in this window]
[in a new window]
  		Figure 4. Time course of the binding of IgG in the plasma of mice immunized with MDA-LDL, LDL, or PBS to MDA-LDL, OxCL, and oxidized cholesterol (Ox-Chol). Plasma samples were obtained before the initial immunization (0 time point) and {approx}1 week after subsequent booster immunizations, ie, after the second (5 weeks), third (7 weeks), fourth (10 weeks), fifth (13 weeks), and seventh (22 weeks) boosters and at the end of the study (24 weeks). Triplicate serum samples from 15 mice immunized with MDA-LDL, 13 mice immunized with LDL, and 11 mice immunized with PBS were diluted 1:1000, and binding of IgG to the plated antigen was determined by CLEIA, as described in Methods. Results are mean±SEM, expressed as RLUs/100 ms. A, To obtain specific binding to MDA-LDL, binding of the plasma IgG to native LDL was subtracted in this assay. At the final time point, binding of IgG antibodies in the plasma of the MDA-LDL–immunized group was significantly higher than that in the LDL and PBS groups (P<0.0001). B, Binding of IgG antibodies to OxCL. Again, the difference between the MDA-LDL group and the LDL and PBS groups at the final time point was highly significant (P<0.0001). C, Binding of IgG antibodies to oxidized cholesterol.

Because autoantibodies to oxidized phospholipids were formed, we investigated whether antibodies were also formed against oxidative products of cholesterol and/or cholesteryl esters, the predominant lipids of LDL. In these assays, we coated microtiter wells with cholesterol or cholesteryl esters and compared their recognition by antibodies to that obtained with the same antigen oxidized by overnight exposure to air. As shown in Figure 4CUp, there was a dramatic rise in antibody binding to oxidized cholesterol. In contrast, binding to unoxidized cholesterol was very low and probably nonspecific (data not shown).

Because the above assays used a high plasma dilution (1:1000) that may not have permitted detection of differences between the LDL and PBS groups with low antibody titers, we also compared the binding to different oxidation products of IgG antibodies in the 3 groups at the end of the intervention period by using lower plasma dilutions (Figure 5Down). At low dilutions, binding to MDA-LDL and Cu-LDL was again significantly greater in the MDA-LDL group than in the other 2 groups (P<0.0001). As shown in the right panels, which use a different scale, binding to both antigens was also slightly but consistently higher in the LDL than in the PBS group (P<0.05 at 1:80 dilution).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 5. Binding of IgG antibodies in plasma of the 3 experimental groups at the end of the study to MDA-LDL, Cu-LDL, and oxidized cholesteryl linoleate. Plasma was diluted as indicated and binding determined by CLEIA, as described in Methods. The right panels show the binding of antibodies in the LDL and PBS groups on a different scale and indicate consistently greater binding in the LDL group. For this assay, a different luminometer was used (Dynex). Thus, luminescence levels (RLUs/100 ms) were lower than those in Figure 4Up.

Binding to oxidized cholesteryl linoleate was also markedly greater in the MDA-LDL group than in the LDL and PBS groups (P<0.0001; Figure 5Up, lower panels), whereas no difference in antibody binding to unoxidized cholesteryl linoleate was detectable (data not shown). Binding to either unoxidized or oxidized cholesteryl oleate was very low and not different between groups (not shown).

In the previous assays, we had used a secondary antibody that detects all IgG antibodies. To investigate the potential role of T helper (Th)1 and Th2 cells, which are involved in the formation of IgG2a and IgG1 antibodies, respectively,43 we determined the binding of these 2 subclasses of antibodies to some of the same antigens. Figure 6Down compares the binding of the 2 IgG isoforms to human MDA-LDL at the start (ie, before immunization) and at the end of the study. Both IgG1 and IgG2a antibodies increased substantially in the MDA-LDL group only, mirroring the increase in total anti–MDA-LDL IgG (Figures 4Up and 5Up). Similar results were also obtained when OxCL was used as the antigen (Figure 7Down). This assay not only compared antibodies at the first and last time points of the study but also followed the IgG1 and IgG2a antibodies over time.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 6. Binding to MDA-LDL of IgG1 and IgG2a antibodies in the plasma of the 3 experimental groups at the start and end of the study. Human MDA-LDL was plated as the antigen and incubated with 1:500 dilutions of plasma obtained before immunization and at the final time point (24 weeks). IgG1 antibody binding to MDA-LDL was detected with a 1:500 dilution of alkaline phosphatase–labeled rat monoclonal antibody to mouse IgG1. IgG2a antibodies were detected with a 1:1000 dilution of alkaline phosphatase–labeled monoclonal rat antibody to mouse IgG2a.



View larger version (29K):
[in this window]
[in a new window]
  		Figure 7. Binding to OxCL of IgCL and IgG2a antibodies in immunized mice. Cardiolipin was dissolved in ethanol, plated as the antigen, and oxidized by exposure to air for 1 hour, as described in Methods. Plasma samples were diluted 1:250 and incubated for 1 hour. Detection of IgG1 and IgG2a isotype antibodies was performed as described in the legend to Figure 6Up. Binding to blank wells incubated with ethanol was subtracted to correct for nonspecific binding to the wells.

We also examined the formation of IgM antibody binding to various oxidative neoepitopes in the 3 experimental groups, because IgM antibodies presumably reflect primarily a T-cell–independent process. In contrast to what we saw for IgG antibodies, binding of IgM antibodies to native LDL showed a small but significant increase of equal magnitude in all 3 groups compared with the preimmune time point (data not shown). Binding of IgM antibodies to MDA-LDL increased to a much greater extent, but surprisingly there was no statistically significant difference between the groups immunized with MDA-LDL, LDL, or PBS (Figure 8ADown). Similarly, there was a marked rise in antibody binding to oxidized cholesterol, but again the rise was similar in all 3 groups (Figure 8BDown). IgM antibody binding to Cu-LDL (Figure 9ADown) and to 4-HNE–LDL (Figure 9BDown) was higher in the MDA-LDL group than in the other 2 groups, albeit the absolute binding was much lower than that seen in Figure 8Down. Surprisingly, binding of IgM antibodies to OxCL (Figure 9CDown) was minimal and did not increase much during the study. This contrasts sharply with the strong IgG response to the same antigen (Figure 4BUp).



View larger version (25K):
[in this window]
[in a new window]
  		Figure 8. Binding of IgM antibodies in the plasma of the 3 experimental groups to human MDA-LDL (A) and oxidized cholesterol (Ox-Chol, B). Antigens were plated at 5 µg/mL and incubated with diluted plasma (1:500), and antibody binding to the plated antigen was detected as described in Methods and in the legend to Figure 4Up, except that alkaline phosphatase–labeled goat anti-mouse IgM (µ-chain specific) was used as the secondary antibody. In these assays, binding to native LDL was not subtracted because IgM antibody binding to native LDL also increased over time in all 3 experimental groups. Data are mean±SEM.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 9. Binding of IgM antibodies in the plasma of the 3 experimental groups to human Cu-LDL, 4-HNE–LDL, and OxCL at the beginning and end of the study. Cu-LDL was a mixture of 2 preparations oxidized for 8 and 16 hours. Murine plasma was diluted 1:500, and IgM antibody binding to the plated antigen was detected with alkaline phosphatase–labeled goat anti-mouse IgM antibody, as described in Methods.

Finally, we sought to test the hypothesis that increased titers of antibodies to oxidation-specific epitopes lead to increased formation of immune complexes with LDL that had undergone minimal modification, because this might affect the removal of such LDL from plasma. IgG and IgM immune complexes on circulating LDL were determined in the terminal plasma samples of all mice. However, no significant differences were found between the 3 groups in either IgG or IgM immune complexes with LDL (for IgG, 23 407±2902, 23 428±4478, and 22 254±1770 RLUs/100 ms; for IgM, 33 829±4752, 28 926±6224, and 28 327±5339 RLUs/100 ms in the MDA-LDL, LDL, and PBS groups, respectively).


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
Atherosclerosis fulfills many of the criteria of a chronic inflammatory process, and substantial histopathological evidence points toward an involvement of humoral and cellular components of the immune system in the disease process. However, the mechanisms involved are poorly understood. While there is little doubt that immune reactions to allografts increase atherogenesis in transplanted arteries,30 31 44 their role in conventional atherosclerosis is controversial, and both beneficial and detrimental consequences of immune modulation have been reported.

The assumption that the immune system may reduce atherogenesis is supported by the increase in atherosclerosis observed in a number of different immunocompromised animal models. For example, elimination of T lymphocytes with monoclonal antibodies resulted in larger proliferative lesions in balloon-catheterized rat aortas.45 Cyclosporine treatment (which suppresses T cells) of hypercholesterolemic mice and rabbits also accelerated atherosclerosis.46 47 MHC class I–deficient C57BL/6 mice, which lack cytolytic T cells and have impaired natural killer cell activity, also developed a 3-fold increase in lesions in the aortic valve region when fed a high-fat diet.48 In contrast, mice lacking either interferon (IFN)-{gamma}44 or the IFN-{gamma} receptor,49 the interleukin-8 receptor,50 or the p55 component of the tumor necrosis factor receptor51 had decreased atherogenesis. ApoE-/- mice crossed with Rag-1 knockout mice showed a 42% decrease in atherosclerosis when fed a regular diet (cholesterol levels of {approx}800 mg/dL) compared with apoE-/- mice. However, the same immunodeficient apoE-/- mice showed no significant decrease in atherosclerosis when fed a high-fat diet that resulted in plasma cholesterol levels of {approx}1800 mg/dL,52 53 suggesting that exceedingly high cholesterol levels overwhelm any modulatory effect of immune deficiency. By analogy, conflicting results have been obtained with interventions inducing or enhancing the humoral immune response to different antigens. For example, immunization with heat shock proteins caused a transitory inflammatory form of lesion in normocholesterolemic rabbits.54 55 In contrast, immunization of hypercholesterolemic WHHL and New Zealand White rabbits with different forms of OxLDL significantly reduced progression of the disease.32 33 34

Mice are a particularly valuable animal model to study the mechanisms by which modulation of the immune system affects atherogenesis, because the murine immune system is well characterized and several immunodeficient strains exist that can be crossed with strains that develop extensive atherosclerosis. In the current article, we demonstrate that continued immunization of LDLR-/- mice with MDA-LDL, a model of OxLDL, yields a highly significant reduction of atherosclerosis in the aortic origin by mechanisms that appear to be independent of the modulation of plasma cholesterol levels. In view of the observations in immunodeficient apoE-/- mice discussed above, it is noteworthy that the extent of atherosclerosis in our experiment was reduced by immunization despite extremely high plasma cholesterol levels.

The current study also provides important mechanistic insights. When we first reported that immunization of WHHL rabbits inhibits the progression of atherogenesis, we proposed 3 potential mechanisms.32 The first of these was the possibility that high titers of antibodies to epitopes of OxLDL enhanced the removal of minimally oxidized (mm) LDL from the circulation. Although it has long been assumed that LDL is oxidized in the arterial wall6 rather than in plasma, where powerful antioxidant protection exists, there is increasing evidence that mmLDL is present in the circulation. For example, plasma LDL of human subjects with manifest cardiovascular disease and ß-VLDL from atherosclerotic rabbits contain significantly more oxidation-specific epitopes then does LDL from controls (References 56 and 5756 57 and J.L.W., unpublished data, 1998). It could therefore be hypothesized that high titers of antibodies to oxidation-specific epitopes enhance formation of immune complexes with mmLDL. This would lead to its rapid removal from the circulation, whereas in the absence of this process, mmLDL might penetrate the vascular wall, where it would be much more susceptible to further oxidation than native LDL. In general, the finding that immunization with MDA-LDL induced a high titer of IgG antibody binding to MDA-lysine epitopes, whereas immunization with native LDL and PBS did not, would not support this hypothesis, although it does not exclude the possibility that this event occurred to some extent in both the MDA-LDL and LDL groups, as discussed below.

In addition to increased anti–MDA-LDL titers, immunization with MDA-LDL also resulted in a strong increase of IgG antibody binding to a variety of other oxidative neoepitopes that one may expect to be present on OxLDL, such as oxidized phospholipids, oxidized cholesterol, and oxidized cholesteryl linoleate. Because MDA modification of LDL should not generate the latter epitopes in principle, this suggests that after immunization the MDA-LDL was rapidly taken up via scavenger receptors of macrophages and that many of these oxidative epitopes were actually generated within macrophages. Because the macrophage is a classic antigen-presenting cell, this would lead to efficient induction of high-titered antibodies. The superiority of antigens (even minimally modified autologous proteins) being taken up by scavenger receptors and inducing T-cell stimulation has previously been established.58 59 Antibodies generated by the above mechanism should bind well to various epitopes of OxLDL.

The humoral immune response to "native" LDL appeared to be far less pronounced than that to MDA-LDL. Nevertheless, it is likely that immunization with LDL also leads to a localized inflammatory condition that resulted in the generation of some mmLDL that would be taken up by macrophages. Indeed, we observed a small but significant increase in IgG antibody binding to MDA-LDL and Cu-LDL in the LDL group compared with the PBS group when the sera were tested at low dilutions (Figure 5Up). A similar small increase in titers was previously reported by Ameli et al.33

The fact that immunization with LDL led to a significant reduction in lesions that was only marginally smaller than that achieved by immunization with MDA-LDL (-36.9% versus -46.3%) suggests that the major beneficial effect of immunization did not result from enhanced removal of mmLDL from the circulation. However, it is possible that relatively small amounts of immunoglobulins formed in the LDL group are sufficient to achieve almost complete removal of mmLDL from the circulation and that a further dramatic increase in titers, as seen in the MDA-LDL group, does not substantially increase benefit. In this case, consumption of antibodies would have markedly reduced the plasma levels of antibodies in the LDL group, whereas the relative impact on the very high titers in the MDA-LDL group would be negligible. Our failure to detect higher antibody titers in mice immunized with LDL could also have resulted from the fact that the antigens used in our CLEIAs did not contain appropriate epitopes formed in vivo, in particular, epitopes that would be destroyed by more extensive oxidation.

The slightly lower plasma cholesterol levels observed in both the LDL and MDA-LDL groups compared with the PBS group would be consistent with the assumption that a small fraction of the LDL was removed from the circulation as a consequence of increased antibody binding to such epitopes. The failure to find differences in circulating immune complexes with plasma LDL does not support this idea, although rapid removal of such immune complexes could have prevented significant differences.

A second postulated mechanism by which high-titered antibodies against oxidation-specific epitopes could have reduced lesion formation would be by enhancing the uptake of OxLDL by macrophages and thereby preventing it from exerting proinflammatory and toxic effects in the vascular wall. We have previously demonstrated that circulating autoantibodies to OxLDL penetrate atherosclerotic lesions and form immune complexes with OxLDL,60 which could enhance the uptake of OxLDL by macrophages via Fc receptors or phagocytosis.14 61 However, the marked discrepancy between antibody titers in the MDA-LDL and LDL groups does not support a major role for this mechanism, either.

A third potential mechanism by which immunization could have affected atherosclerosis is through cell-mediated immune mechanisms. Atherosclerotic lesions of mice resemble those of other species in that they contain both CD4+ and CD8+ cells but few if any B cells.62 63 Lesions also contain large numbers of antigen-presenting cells, such as macrophages, bearing MHC class II molecules on their surfaces, which colocalize with IFN-{gamma} and interleukins.29 49 Oxidation-specific epitopes are abundant in lesions, both extracellularly and within macrophages.15 28 It is therefore likely that modified peptides of apoB derived from OxLDL are candidates for antigen presentation by classic human leukocyte antigen A (HLA) class II mechanisms. It is also likely that oxidized lipids or lipid-protein adducts are presented as antigens, via mechanisms such as the recently described CD1 molecule.64 All cell-mediated immune mechanisms presumably result from initial T-cell recognition of antigen,65 66 and indeed, 10% of CD4+ T cells cloned from human lesions specifically proliferated in response to OxLDL in an HLA class II–restricted manner.67

Once antigen-activated, T cells secrete a variety of cytokines that could have a profound effect on the progression of the atherogenic process. At present, we do not know whether differences in T-cell populations, such as cytolytic lymphocytes or natural killer cells, or differences in their secretory products exist between the MDA-LDL and LDL groups on 1 side and the PBS group on the other. Th1 and Th2 cells are known to have different patterns of cytokine secretion and have been postulated to play different roles in promoting inflammatory conditions, as well as in cross-regulation of each other's activity.43 68 69 Therefore, one could assume that a shift from 1 T-cell subset to another may have influenced the balance of proatherogenic and antiatherogenic factors secreted in lesions. Such shifts during the course of atherosclerosis have recently been described in apoE-/- mice.70 In response to immunization with MDA-LDL, there appeared to be a generalized increase in both Th1 and Th2 subtypes (judged by rises in both IgG1 and IgG2a antibodies). However, no comparable increase in IgG1 or IgG2a was seen in the LDL group (which also received Freund's adjuvant). Furthermore, it is unknown whether the differences in the level of plasma antibodies to the epitopes of OxLDL accurately reflect differences in Th1 and Th2 cell numbers or activity in the lesions. Only direct studies of the various lymphocyte populations in lesions, in lymph nodes draining atherosclerotic arteries, and of memory cells at other sites will determine their relative importance. Adoptive transfer experiments could also establish the role of cell-mediated immunity in the protective effect.

We have previously reported that IgM antibodies to oxidation-specific epitopes are prominent in atherosclerotic mice15 16 28 and human subjects with clinically manifest coronary heart disease. The generation of these antibodies is thought to be largely T-cell independent. In the current study there was a steady rise in IgM antibodies to a variety of oxidation-specific epitopes, including oxidized lipids. Because the rise was similar in all 3 experimental groups, it is not likely that IgM directly contributed to the protective effect observed, although they could participate in the natural history of lesion development.

The fact that a significant reduction in lesion size was observed in cross sections of the aortic origin, but not in the percentage of aortic surface stained with Sudan IV, suggests that immunization inhibited the growth of lesions but did not significantly impact lesion initiation. In LDLR-/- and apoE-/- mice, lesions first appear at a few very consistent predilection sites in the arch and near the orifices of the mesenteric arteries. In our experience, lesions at these sites undergo considerable intimal thickening before they begin to cover a larger percentage of the aortic tree. Later, horizontal spreading of lesions occurs mostly in the abdominal aorta and arch, whereas the thoracic aorta shows only isolated lesions.15 28 41 42 It could therefore be assumed that the limited lesions we observed had not reached a stage when they would rapidly spread. Consequently, in this setting, determination of the surface area would not be a very sensitive parameter to detect a beneficial effect of immunization. The conclusion that intervention affected progression rather than initiation of lesion formation is also supported by our previous immunization study in WHHL rabbits.32 In that study, immunization of young animals failed to yield a significant effect on early stages of atherogenesis in the aorta, whereas immunization of older animals with more extensive initial atherosclerosis reduced lesion formation by 35%.


*    Acknowledgments
 
These studies were supported by grants HL14197 and HL5989 (La Jolla Specialized Center of Research in Molecular Medicine and Atherosclerosis; to J.L.W. and W.P.) from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md; grant HL57505 (to J.L.W.); and a grant from the German Academic Exchange Service (DAAD) (to S.F.). The current article is part of Stefan Freigang's Medical Doctorate Thesis, Humboldt University, Berlin, Germany. We thank Florencia Casanada, Joe Juliano, and Jennifer Pattison for excellent technical assistance.

Received May 26, 1998; accepted July 29, 1998.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Palinski W, Rosenfeld ME, Ylä-Herttuala S, Gurtner GC, Socher SA, Butler S, Parthasarathy S, Carew TE, Steinberg D, Witztum JL. Low density lipoprotein undergoes oxidative modification in vivo. Proc Natl Acad Sci U S A. 1989;86:1372–1376.[Abstract/Free Full Text]
  2. Haberland ME, Fong D, Cheng L. Malondialdehyde-altered protein occurs in atheroma of Watanabe heritable hyperlipidemic rabbits. Science. 1988;241:215–241.[Abstract/Free Full Text]
  3. Boyd HC, Gown AM, Wolfbauer G, Chait A. Direct evidence for a protein recognized by a monoclonal antibody against oxidatively modified LDL in atherosclerotic lesions from a Watanabe Heritable Hyperlipidemic rabbit. Am J Pathol. 1989;135:815–826.[Abstract]
  4. Rosenfeld ME, Palinski W, Ylä-Herttuala S, Butler S, Witztum JL. Distribution of oxidation-specific lipid-protein adducts and apolipoprotein B in atherosclerotic lesions of varying severity from WHHL rabbits. Arteriosclerosis. 1990;10:336–349.[Abstract/Free Full Text]
  5. Hammer A, Kager G, Dohr G, Rabl H, Ghassempur I, Jürgens G. Generation, characterization, and histochemical application of monoclonal antibodies selectively recognizing oxidatively modified apoB-containing serum lipoproteins. Arterioscler Thromb Vasc Biol. 1995;15:704–713.[Abstract/Free Full Text]
  6. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low density lipoprotein that increase its atherogenicity. N Engl J Med. 1989;320:915–924.[Medline] [Order article via Infotrieve]
  7. Witztum JL. The oxidation hypothesis of atherosclerosis. Lancet. 1994;344:793–795.[Medline] [Order article via Infotrieve]
  8. Berliner JA, Navab M, Fogelman AM, Frank JS, Demer LL, Edwards PA, Watson AD, Lusis AJ. Atherosclerosis: basic mechanisms: oxidation, inflammation, and genetics. Circulation. 1995;91:2488–2496.[Abstract/Free Full Text]
  9. Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of 4-hydroxynonenal, malondialdehyde and related aldehydes. Free Radic Biol Chem. 1991;11:81–128.
  10. Steinbrecher UP, Fisher M, Witztum JL, Curtiss LK. Immunogenicity of homologous low density lipoprotein after methylation, ethylation, acetylation or carbamylation: generation of antibodies specific for derivatized lysine. J Lipid Res. 1984;25:1109–1116.[Abstract]
  11. Curtiss LK, Witztum JL. A novel method for generating region specific monoclonal antibodies to modified proteins: application to the identification of human glucosylated low density lipoproteins. J Clin Invest. 1983;72:1427–1438.
  12. Palinski W, Ylä-Herttuala S, Rosenfeld ME, Butler S, Socher SA, Parthasarathy S, Curtiss LK, Witztum JL. Antisera and monoclonal antibodies specific for epitopes generated during the oxidative modification of low density lipoprotein. Arteriosclerosis. 1990;10:325–335.[Abstract/Free Full Text]
  13. Parums DV, Brown DL, Mitchinson MJ. Serum antibodies to oxidized low-density lipoprotein and ceroid in chronic periaortitis. Arch Pathol Lab Med. 1990;114:383–387.[Medline] [Order article via Infotrieve]
  14. Orekhov AN, Tertov VV, Kabakov AE, Adamova IYu, Pokrovsky SN, Smirnov VN. Autoantibodies against modified low density lipoprotein: nonlipid factor of blood plasma that stimulates foam cell formation. Arterioscler Thromb. 1991;11:316–326.[Abstract/Free Full Text]
  15. Palinski W, Ord V, Plump AS, Breslow JL, Steinberg D, Witztum JL. ApoE-deficient mice are a model of lipoprotein oxidation in atherogenesis: demonstration of oxidation-specific epitopes in lesions and high titers of autoantibodies to malondialdehyde-lysine in serum. Arterioscler Thromb. 1994;14:605–616.[Abstract/Free Full Text]
  16. Palinski W, Hörkkö S, Miller E, Steinbrecher UP, Powell HC, Curtiss LK, Witztum JL. Cloning of monoclonal autoantibodies to epitopes of oxidized lipoproteins from apoE-deficient mice: demonstration of epitopes of oxidized low density lipoprotein in human plasma. J Clin Invest. 1996;98:800–814.[Medline] [Order article via Infotrieve]
  17. Witztum JL, Palinski W. Autoimmunity to oxidized lipoproteins. In: Hansson GK, Libby P, eds. Immune Functions of the Vessel Wall. Amsterdam, Netherlands: Harwood Academic Publishers; 1996:159–172.
  18. Salonen JT, Ylä-Herttuala S, Yamamoto R, Butler S, Korpela H, Salonen R, Nyyssönen K, Palinski W, Witztum JL. Autoantibody against oxidised LDL and progression of carotid atherosclerosis. Lancet. 1992;339:883–887.[Medline] [Order article via Infotrieve]
  19. Maggi E, Finardi G, Poli M, Bollati P, Filipponi M, Stefano PL, Paolini G, Grossi A, Clot P, Albano E, Bellomo G. Specificity of autoantibodies against oxidized LDL as an additional marker for atherosclerotic risk. Coron Artery Dis. 1993;4:1119–1122.[Medline] [Order article via Infotrieve]
  20. Virella G, Virella I, Leman RB, Pryor MB, Lopes-Virella MF. Antioxidized low density lipoprotein antibodies in patients with coronary heart disease and normal healthy volunteers. Int J Clin Lab Res. 1993;23:95–101.[Medline] [Order article via Infotrieve]
  21. Branch DW, Mitchell MD, Miller E, Palinski W, Witztum JL. Pre-eclampsia and serum antibodies to oxidised low density lipoprotein. Lancet. 1994;343:645–646.[Medline] [Order article via Infotrieve]
  22. Armstrong VW, Wieland E, Diedrich F, Renner A, Rath W, Kreuzer H, Kuhn W, Oellerich M. Serum antibodies to oxidised low-density lipoprotein in pre-eclampsia and coronary heart disease. Lancet. 1994;343:1570.[Medline] [Order article via Infotrieve]
  23. Maggi E, Chiesa R, Melissano G, Castellano R, Astore D, Grossi A, Finardi G, Bellomo G. LDL oxidation in patients with severe carotid atherosclerosis: a study of in vitro and in vivo oxidation markers. Arterioscler Thromb. 1994;14:1892–1899.[Abstract/Free Full Text]
  24. Puurunen M, Mänttäri M, Manninen V, Tenkanen L, Alfthan G, Ehnholm C, Vaarala O, Aho K, Palosuo T. Antibody against oxidized low-density lipoprotein predicting myocardial infarction. Arch Intern Med. 1994;154:2605–2609.[Abstract]
  25. Bellomo G, Maggi E, Poli M, Agosta FG, Bollati P, Finardi G. Autoantibodies against oxidatively modified low-density lipoproteins in NIDDM. Diabetes. 1995;44:60–66.[Abstract]
  26. Bergmark C, Wu R, de Faire U, Lefvert AK, Swedenborg J. Patients with early-onset peripheral vascular disease have increased levels of autoantibodies against oxidized LDL. Arterioscler Thromb Vasc Biol. 1995;15:441–445.[Abstract/Free Full Text]
  27. Heitzer T, Ylä-Herttuala S, Luoma J, Kurz S, Münzel T, Just H, Olschewski M, Drexler H. Cigarette smoking potentiates endothelial dysfunction of forearm resistance vessels in patients with hypercholesterolemia: role of oxidized LDL. Circulation. 1996;93:1346–1353.[Abstract/Free Full Text]
  28. Palinski W, Tangirala RK, Miller E, Young SG, Witztum JL. Increased autoantibody titers against epitopes of oxidized low density lipoprotein in LDL receptor-deficient mice with increased atherosclerosis. Arterioscler Thromb Vasc Biol. 1995;15:1569–1576.[Abstract/Free Full Text]
  29. Libby P, Hansson GK. Involvement of the immune system in human atherogenesis: current knowledge and unanswered questions. Lab Invest. 1991;64:5–15.[Medline] [Order article via Infotrieve]
  30. Tanaka H, Swanson SJ, Sukhova G, Schoen FJ, Libby P. Smooth muscle cells of the coronary arterial tunica media express tumor necrosis factor-{alpha} and proliferate during acute rejection of rabbit cardiac allografts. Am J Pathol. 1995;147:617–626.[Abstract]
  31. Shi C, Lee WS, He Q, Zhang D, Fletcher DL Jr, Newell JB, Haber E. Immunologic basis of transplant-associated arteriosclerosis. Proc Natl Acad Sci U S A. 1996;93:4051–4056.[Abstract/Free Full Text]
  32. Palinski W, Miller E, Witztum JL. Immunization of LDL receptor-deficient rabbits with homologous malondialdehyde-modified LDL reduces atherogenesis. Proc Natl Acad Sci U S A. 1995;92:821–825.[Abstract/Free Full Text]
  33. Ameli S, Hultgårdth-Nilsson A, Regnström J, Calara F, Yano J, Cercek B, Shah PK, Nilsson J. Effect of immunization with homologous LDL and oxidized LDL on early atherosclerosis in hypercholesterolemic rabbits. Arterioscler Thromb Vasc Biol. 1996;16:1074–1079.[Abstract/Free Full Text]
  34. Nilsson J, Calara F, Regnström J, Hultgårdth-Nilsson A, Ameli S, Cercek B, Shah PK. Immunization with homologous oxidized low density lipoprotein reduces neointimal formation after balloon injury in hypercholesterolemic rabbits. J Am Coll Cardiol. 1997;30:1886–1891.[Abstract]
  35. Ishibashi S, Brown MS, Goldstein JL, Gerard RD, Hammer RE, Herz J. Hypercholesterolemia in low density lipoprotein receptor knockout mice and its reversal by adenovirus-mediated gene delivery. J Clin Invest. 1993;92:883–893.
  36. Habeeb AFSA. Chemical evaluation of conformational differences in native and chemically modified proteins by trinitrobenzenesulfonic acid. Biochim Biophys Acta. 1966;115:440–454.[Medline] [Order article via Infotrieve]
  37. Ylä-Herttuala S, Palinski W, Rosenfeld ME, Parthasarathy S, Carew TE, Butler S, Witztum JL, Steinberg D. Evidence for the presence of oxidatively modified low density lipoprotein in atherosclerotic lesions of rabbit and man. J Clin Invest. 1989;84:1086–1095.
  38. Hörkkö S, Miller E, Dudl E, Reaven PD, Zvaifler NJ, Terkeltaub R, Pierangeli SS, Curtiss LK, Branch W, Palinski W, Witztum JL. Antiphospholipid antibodies are directed against epitopes of oxidized phospholipids: recognition of cardiolipin by monoclonal antibodies to epitopes of oxidized low density lipoprotein. J Clin Invest. 1996;98:815–825.[Medline] [Order article via Infotrieve]
  39. Hörkkö S, Miller E, Branch DW, Palinski W, Witztum JL. The epitopes for some antiphospholipid antibodies are adducts of oxidized phospholipid and ß2-glycoprotein 1 (and other proteins). Proc Natl Acad Sci U S A. 1997;94:10356–10361.[Abstract/Free Full Text]
  40. Iuliano L, Praticò D, Ferro D, Pittoni V, Valesini V, Lawson J, FitzGerald GA, Violi F. Enhanced lipid peroxidation in patients positive for antiphospholipid antibodies. Blood. 1997;90:3931–3935.[Abstract/Free Full Text]
  41. Tangirala RK, Rubin EM, Palinski W. Quantitation of atherosclerosis in murine models: correlation between lesions in the aortic origin and in the entire aorta, and differences in the extent of lesions between sexes in LDL receptor-deficient and apoprotein E-deficient mice. J Lipid Res. 1995;36:2320–2328.[Abstract]
  42. Reaven P, Merat S, Casanada F, Sutphin M, Palinski W. Effect of streptozotocin induced hyperglycemia on lipid profiles, formation of advanced glycation endproducts in lesions, and extent of atherosclerosis in LDL receptor–deficient mice. Arterioscler Thromb Vasc Biol. 1997;17:2250–2256.[Abstract/Free Full Text]
  43. Mosmann TR, Coffman RL. Heterogeneity of cytokine secretion patterns and functions of helper T-cells. Adv Immunol. 1989;46:111–147.[Medline] [Order article via Infotrieve]
  44. Nagano H, Mitchell RN, Taylor MK, Hasegawa S, Tilney NL, Libby P. Interferon-{gamma} deficiency prevents coronary arteriosclerosis but not myocardial rejection in transplanted mouse hearts. J Clin Invest. 1997;100:550–557.[Medline] [Order article via Infotrieve]
  45. Hansson GK, Holm J, Holm S, Fotev Z, Hedric HJ, Fingerle J. T lymphocytes inhibit the vascular response to injury. Proc Natl Acad Sci U S A. 1991;88:10530–10534.[Abstract/Free Full Text]
  46. Emeson EE, Shen ML. Accelerated atherosclerosis in hyperlipidemic C57BL/6 mice treated with cyclosporin A. Am J Pathol. 1993;142:1906–1915.[Abstract]
  47. 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.
  48. Fyfe AI, Qiao JH, Lusis AJ. Immune deficient mice develop typical atherosclerotic fatty streaks when fed an atherogenic diet. J Clin Invest. 1994;94:2516–2520.
  49. Gupta S, Pablo AM, Jiang X, Wang N, Tall AR, Schindler C. IFN-{gamma} potentiates atherosclerosis in ApoE knock-out mice. J Clin Invest. 1997;99:2752–2761.[Medline] [Order article via Infotrieve]
  50. Boisvert WA, Santiago R, Curtiss LK, Terkeltaub RA. A leukocyte homologue of the IL-8 receptor CXCR-2 mediates the accumulation of macrophages in atherosclerotic lesions of LDL receptor-deficient mice. J Clin Invest. 1998;101:353–363.[Medline]