|Home | About | Journals | Submit | Contact Us | Français|
Cardiovascular disease risk is increased in individuals suffering from systemic lupus erythematosus (SLE). Understanding the mechanism(s) of SLE-accelerated atherosclerosis is critical for the development of effective therapies. Our laboratory previously demonstrated that radiation chimeras of SLE-susceptible B6.Sle1.2.3 and low density lipoprotein receptor (LDLr)−/−mice have augmented atherosclerosis which is associated with increased T cell burden and activation in the lesion. The goals of this study were to further define specific immune mechanisms mediating accelerated atherosclerosis and to determine whether the gene interval Sle3, which is linked to lupus-associated T cell dysregulation, was sufficient to modulate atherogenesis. We transferred B6.Sle3 or C57Bl/6-derived bone marrow cells into lethally irradiated LDLr−/− mice (hereafter referred to as LDLr.Sle3 and LDLr.B6, respectively). Sixteen weeks after transplantation, the mice were placed on a Western-type diet for 8 weeks. Our analyses revealed that LDLr.Sle3 mice had increased auto-antibody production against dsDNA and cardiolipin compared to LDLr.B6 controls. We also found an increase in atherosclerosis associated oxLDL antibodies. Antibody isotypes and serum cytokine analysis suggested that the humoral immune response in LDLr.Sle mice was skewed toward a Th2 phenotype. This is consistent with lupus-associated immune dysregulation. Additionally, LDLr.Sle3 mice had decreased serum cholesterol and triglyceride levels. However, there was no difference in lesion area or cellular composition of lesions between the two groups. These data demonstrate that, despite no changes in lesion area, transfer of Sle3-associated T cell dysregulation alone to LDLr-deficient mice is sufficient to decrease serum cholesterol and to exacerbate humoral immune responses that are frequently associated with atherosclerosis.
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by the presence of autoantibodies against many autoantigens such as histones, dsDNA, and oxLDL (1, 2). With the development of new treatments to alleviate the often fatal complications of SLE, patient survival has increased. This has led to the discovery that SLE is associated with an increase in cardiovascular disease (CVD) events(3). Most impressive are studies showing that the incidence of CVD is even greater for women, as pre-menopausal women are 50 times more likely to develop CVD (4).
Atherosclerosis, one of the most common processes leading to the development of CVD, is traditionally associated with hypertension, dyslipidemia, and hypercholesterolemia. Recent evidence has shown that the immune system also plays a role in this disease process as macrophages, T cells, B cells, autoantibodies and autoantigens have all been found within the atherosclerotic lesion (5–7). The involvement of the immune system in atherosclerosis is especially evident in SLE as traditional risk factors alone cannot account for the prevalence of CVD in these patients (8–11). Therefore, in order to reduce the risk of SLE-associated CVD, it is necessary to define the immune mechanisms that mediate atherosclerosis in the face of autoimmunity.
Until recently, understanding the process of SLE-mediated acceleration of atherosclerosis using animal models was challenging as mouse strains susceptible to lupus are typically resistant to developing atherosclerosis. Through linkage analyses, Morel et al., identified three major genomic intervals linked to lupus susceptibility in the NZM2410 mouse strain (12). Using these three chromosomal intervals, termed Sle1, Sle2 and Sle3, the investigators made a series of single, bi-, and triple congenic mice on the atherosclerosis susceptible C57Bl/6 background (13, 14). Sle1 is associated with chronic lymphocyte activation and anti-nuclear antibodies production (15, 16). Sle2 is thought to lower the activation threshold of B cells leading to B cell hyperactivity, B-1 B cell expansion and polyclonal IgM production (17). Sle3 mediates CD4+ T and antigen presenting cell hyperactivity and is associated with decreased activation induced cell death and an elevated CD4:CD8 T cell ratio (18–20). While having one or two intervals can confer varying symptoms associated with lupus, mice with all three intervals display a fully penetrant lupus phenotype similar to the disease in humans.
Our laboratory recently created an animal model of SLE-accelerated atherosclerosis by demonstrating that transfer of lupus susceptibility by bone marrow transplantation increases atherosclerosis in low density lipoprotein receptor-deficient (LDLr−/−) mice (21). The increase in atherosclerosis was later determined to be independent of diet (22) and was accompanied by a three-fold increase in CD4+ T cell burden within the atherosclerotic lesion area (21, 22). CD4+ T cells from the SLE-susceptible mice also displayed higher expression of activation markers such as CD69 (21) and CD40L (22). Additionally, it has been observed that in humans, plaque stability is inversely associated with T cell burden in humans (23, 24). These data led us to hypothesize that T cell dysregulation, which is known to be mediated by the lupus susceptibility interval Sle3, may facilitate the accelerated atherosclerosis seen in mice susceptible to lupus. In the current study, we examined whether Sle3 is sufficient to accelerate atherosclerosis.
C57Bl/6J (B6) and B6.129S7-Ldlrtm1Her/J (LDLr−/−) mice were originally purchased from The Jackson Laboratory (Bar Harbor, ME) and maintained in the Vanderbilt University animal care facility. All procedures were approved by the Vanderbilt Institutional Animal Care and Use Committee. The B6.NZMc7 (B6.Sle3) mice, a generous gift from Edward Wakeland at the University of Texas Southwestern Medical Center, are C57Bl/6 mice congenic for the NZM2410-derived chromosome 7 lupus susceptibility interval. This single congenic strain has been described previously (18–20).
Transfer of bone marrow was accomplished by bone marrow transplantation as described previously (21). Female LDLr−/− mice received either C57Bl/6 or B6.Sle3 bone marrow, hereafter referred to as LDLr.B6 and LDLr.Sle3, respectively. Sixteen weeks after transplantation, mice were placed on a high-fat Western diet (21% milk fat, 0.15% cholesterol) for eight weeks. Mice were then sacrificed and analyzed for the degree of atherosclerosis and the presence and severity of symptoms associated with SLE.
Total serum cholesterol and triglyceride were measured in mice fasted for at least four hours using a colorimetric assay as described previously (25). Lipoprotein distribution was determined by using FPLC.
Serum cytokine levels were analyzed using the Milliplex Mouse Cytokine/Chemokine kit according to manufacturer’s protocol and detected using Luminex® xMAP® (Millipore, Billerica, MA).
Serum titers of dsDNA were measured according to the method of Shivakumar et al. (26). Anti-oxLDL antibodies were measured as described previously (27). Serum antibody titers against β2-glycoprotein I (β2-GPI) were measured by coating a 96-well Maxisorb plate with 10 μg/ml of purified β2GPI in 1% bovine serum albumin (1% BSA)/PBS overnight. Plates were blocked in 1% BSA/PBS for two hours at room temperature. Mouse serum was added at a dilution between 1:500 and 1:5000 and incubated overnight at 4°C. Plates were washed with 0.5% Tween-20/PBS (PBS-T) and incubated with biotin-conjugated goat anti-mouse Ig(H+L) (SouthernBiotech, Birmingham, AL) for 45 minutes at room temperature then incubated with avidin-peroxidase for 30 minutes at room temperature. Plates were then washed with PBS-T and developed using TMB substrate (BD Bioscience). Anti-β2-GPI immunoglobulin isotype ELISAs were performed as described above using a biotin-conjugated goat anti-mouse IgG1, IgG2A or IgM (SouthernBiotech) secondary antibody. Cardiolipin antibody titers were determined by coating a 96-well Maxisorb plate with cardiolipin (Sigma-Aldrich) (15μg/ml in 95% ethanol). The cardiolipin ELISA was then conducted as described above for β2-GPI. Serum was diluted 1:1000, 1:500 and 1:5000 for oxLDL, β2-GPI and cardiolipin ELISAs, respectively.
Staining for macrophages (MOMA-2) and CD4+ T cells was performed as described previously (21, 22). Cells were visualized and staining quantified using Image-Pro Plus software (Media Cybernetics, Bethesda, MD).
For flow cytometric analyses, spleens were removed and processed through a 0.70 μm mesh screen. Cells were counted, resuspended in 4% fetal bovine serum (FBS) in PBS with 0.5% sodium azide, and incubated with appropriate antibodies for 40 minutes at 4°C. Cells were then washed and analyzed using a 5-Laser BD LSRII flow cytometer (BD Bioscence) and FacsDiva software (BD Bioscience). The following antibodies were used: TCRβ (H57–597), CD8α (53–6.7), CD4 (GK1.5), CD69 (H1.2F3), B220 (RA3-6B2), NK1.1 (PK136), CD44 (IM7), CD40L (MR1), CD11b (M1/70), CD40 (1C10), CD80 (16-10A1) and CD86 (GL1) (all purchased from BD Bioscience).
Statistical analyses were conducted using PRISM 5.0 software (GraphPad Software Inc., La Jolla, CA). For data with a normal Gaussian distribution, a Student’s t-test was used to calculate significant differences between groups. For data not normally distributed, a Mann-Whitney test was performed. A p value of < 0.05 was considered significant.
B6.Sle3 mice are known to produce antibodies against dsDNA (14). To confirm the hematopoietic transfer of the lupus phenotype into LDLr−/− hosts, dsDNA antibody titers were measured in serum collected at time of sacrifice. As expected, LDLr.Sle3 mice had higher dsDNA antibody titers compared to control (Figure 1A). Examination of spleen weights at the time of sacrifice showed that LDLr.Sle3 mice had increased spleen:body weight ratios compared to LDLr.B6 mice (Figure 1B).
The congenic B6.Sle3 mouse model is additionally associated with T cell hyperactivity, increased CD4/CD8 T cell ratios, and hyperstimulatory antigen presenting cells in the absence of spleen size differences when compared to B6 controls (18–20). We sought to determine if these differences in splenic cell populations could also be observed when the NZM2410-derived Sle3 congenic interval is transferred hematopoietically to atherosclerosis susceptible LDLr−/− mice fed a western-type diet. Spleen cells were incubated with the panel of fluorophore-conjugated antibodies as outlined in the “Methods” section and cell populations were analyzed by flow cytometry. Flow cytometric analyses showed no significant differences in B cell, CD8+ T cell, macrophage, dendritic cell or NK cell numbers in the spleens of LDLr.B6 and LDLr.Sle3 mice (Figure 1C). However, we did observe a variable 2-fold increase in CD4+ T cell numbers and CD4:CD8 T cell ratios. The increase in CD4+ T cells was also accompanied by an increase in the activation marker CD40L and a trend toward increased percentages of CD4+CD69+ and CD4+CD44+ T cells in LDLr.Sle3 mice (Figure 1D–E).
Sixteen weeks after transplantation, mice were placed on a western diet containing 21% milk fat and 0.15% cholesterol for eight weeks. After eight weeks, mice were sacrificed and analyzed for severity of atherosclerosis. Examination of atherosclerotic lesion area in the proximal aorta of LDLr.B6 and LDLr.Sle3 by oil-red-O staining revealed no statistically significant difference in lesion area between the two groups (Figure 2A). In addition, the cellular composition of the atherosclerotic plaque, as assessed by Moma-2 and CD4 staining, was similar in both lupus-susceptible and control animals (Figure 2B and C).
Interestingly, as observed in our previous studies with LDLr.Sle1.2.3 mice, measurement of serum cholesterol and triglyceride levels revealed a significant decrease in cholesterol in LDLr.Sle3 mice compared to LDLr.B6 mice (Figure 3A). In addition, although not statistically significant, there was a trend for lower serum triglyceride levels in LDLr.Sle3 mice compared to controls (Figure 3B). FPLC analysis revealed that the 20% decrease in serum cholesterol in LDLr.Sle3 mice was due to a decrease in the very low density lipoprotein (VLDL) and low density lipoprotein (LDL) cholesterol fractions (Figure 3C). Overall, these results show that while there was no difference in lesion area and cellular composition of plaques between the two groups, the LDLr.Sle3 mice did have lower cholesterol and triglyceride levels than control animals.
We examined total immunoglobulin levels against the anti-phospholipid antibody cofactor β2-GPI and the phospholipid cardiolipin. Measurements of antibodies against both antigens are indicative of anti-phospholipid antibody titers. We observed a trend toward increased anti-β2-GPI (Figure 4A) and anti-cardiolipin (Figure 4B) immunoglobulin levels in LDLr.Sle3 mice although neither reached statistical significance. Antibody measurements for oxLDL also revealed the same trend, with LDLr.Sle3 mice producing more antibodies than control (Figure 4C).
Evaluation of anti-β2-GPI and anti-cardiolipin immunoglobulin isotype titers demonstrated significantly higher IgG1 and IgM, but not IgG2c, titers in the sera of LDLr.Sle3 mice compared to control mice. Calculation of the IgG1 (Th2) to IgG2c (Th1) isotype ratio demonstrated that LDLr.Sle3 mice had an increased IgG1/IgG2c ratio (Table 1). We then went on to analyze the levels of IFN-γ and IL-4 in the sera of these mice and found significantly less IFN-γ levels (p=0.036) in LDLr.Sle3 mice (35.22pg/ml; n=8) compared to LDLr.B6 controls (66.33pg/ml; n=8). IL-4 was not different between the two groups (data not shown). Taken together, these results show that the LDLr.Sle3 mice have increased auto-antibody production compared to control animals and that hematopoietic transfer of the Sle3 lupus-susceptibility loci is sufficient to exacerbate humoral immune responses against atherosclerosis-associated antigens, such as β2-GPI and oxLDL.
It is well known that individuals with SLE have an increased risk of developing CVD. Our laboratory (21, 22), and others (28), have shown that accelerated atherosclerosis can occur following hematopoietic transfer of lupus to LDLr−/− mice and that immune dysregulation and chronic inflammation can modulate atherosclerosis. However, the exact mechanisms mediating accelerated vascular disease in lupus are yet to be determined.
In an effort to elucidate possible mechanisms of SLE-accelerated atherosclerosis, we examined whether the Sle3 lupus-susceptibility locus alone was sufficient to increase vascular disease in LDLr−/− mice. Sle3 was found to mediate T cell and antigen presenting cell hyperactivity (18–20). In the present study, we hypothesized that T cell dysregulation, mediated by Sle3, is sufficient to accelerate atherosclerosis. However, we found no differences in proximal aortic lesion sizes or cellular composition between LDLr.B6 and LDLr.Sle3. Our results suggest that just as having one lupus susceptibility interval is not sufficient to induce fully penetrant lupus disease, 2 or more lupus susceptibility loci may be necessary to accelerate atherosclerosis in mice.
Patients with SLE produce large amounts of autoantibodies, including antibodies against atherosclerotic antigens (29, 30). In our model, analyses of humoral immune responses against such antigens demonstrated an increase in antibody production in the absence of increased atherosclerosis. When isotype specific immunoglobulin levels against β2-GPI and cardiolipin were measured, we found increased IgG1 and IgM antibodies along with an increased IgG1/IgG2c ratio in LDLr.Sle3 mice compared to controls. These results are congruent with our previous study using LDLr.Sle1.2.3. mice (22) and demonstrate that while Sle3 is not able to mediate accelerated atherosclerosis, it is sufficient to induce humoral responses similar to those seen in the triple congenic model.
While antibody titers against atherosclerosis-associated antigens are traditionally thought to be a biomarker for cardiovascular disease risk (31, 32), reports suggest such responses can be pro- or anti-atherogenic (29). Antibodies of the IgG isotype, specific for oxLDL/ β2-GPI immune complexes are postulated to be proatherogenic as they may facilitate uptake of immune complexes through Fc receptors (33). However, other reports show that anti-oxLDL (mainly IgM) and anti-cardiolipin could potentially be anti-atherogenic (34, 35). In general, both IgM and IgG1 antibodies are thought to be atheroprotective. Additionally, IgG1 antibodies indicate a Th2 immune response as IL-4 facilitates B cell class switching to the IgG1 isotype. Conversely, Th1 cells, through their production of IFN-γ, facilitate IgG2c class switching, and IFN-γ is thought to be pro-atherogenic (36–38). Given these reports, our results present an interesting outcome. In this study, we find that although LDLr.B6 mice have an increased Th1 antibody response and LDLr.Sle3 mice have increased atheroprotective antibodies, there are no differences in lesion area between the groups. Therefore, despite having more possibly atheroprotective antibodies, the atherosclerotic lesions develop similarly in LDLr.Sle3 mice compared to LDLr.B6 mice. Thus it is possible that that although the Sle3 gene locus is not enough to push the animals toward increased atherosclerosis, it does decrease the protective phenotype or functions of atherosclerosis-associated antibodies. It is possible that addition of the Sle1 or Sle2 loci would be enough to increase autoimmune dysregulation and increase atherosclerosis in this model. Obviously this hypothesis is the focus of future investigations by our laboratory and will allow us to ascertain how epistatic interaction between these loci may affect CVD.
Parallel with our previous studies, we also observed decreased serum cholesterol and triglyceride levels in LDLr.Sle3 mice compared to control animals. (21, 22) This is an interesting observation since it suggests that either genes encoded on the Sle3 chromosomal interval are playing a role in cholesterol homeostasis in the LDLr.Sle3 mice, or that increased anti-oxLDL and anti-phospholipid antibody levels in the lupus-susceptible mice increase uptake and/or clearance of circulating lipoproteins. We hypothesize the latter because cholesterol and triglyceride levels are similar between LDLr.B6 and LDLr.Sle3 mice before the initiation of Western diet and the appearance of dsDNA antibodies; the first biomarker of lupus-associated immune dysregulation. Again, we find that despite lower cholesterol levels, our LDLr.Sle3 mice have similar atherosclerotic lesion area compared to controls. These data lend further support to the hypothesis that autoimmunity rather than traditional risk factors, such as elevated cholesterol levels, is the primary mediator of accelerated atherosclerosis in the autoimmune disease.
Previously, our laboratory reported a three-fold increase in CD4+ T cell burden in the atherosclerotic plaques of LDLr.Sle1.2.3 (21, 22). This led us to hypothesize that the increase in vascular disease in these mice may be mediated primarily by T cell hyperactivity. Our recent findings suggest that while there is evidence of a hyperactive T cell phenotype in LDLr.Sle3 mice (Figure 1C–E), Sle3 by itself is not enough to initiate CD4+ T cell infiltration into atherosclerotic plaques. Furthermore, the data also indicate that T cell accumulation in lesions may be one driving force of lesion progression in LDLr.Sle1.2.3 mice as the LDLr.B6 and LDLr.Sle3 mice had similar amounts of T cell percentages within the lesion. Indeed this is a plausible hypothesis as Zhou et al. found that transfer of CD4+ T cells into ApoE−/−scid/scid mice drastically increased lesion formation and that these T cells also homed to the atherosclerotic lesion area (39, 40). Further studies examining the kinetics of T cell migration into lesions are necessary to determine if this is the case.
In conclusion, we have reported that transfer of the lupus susceptibility interval Sle3 to LDLr−/− mice is not sufficient to accelerate the progression of atherosclerosis. It can, however, mediate antibody production against dsDNA and antigens typically associated with atherosclerosis, such as phospholipids and oxLDL. Additionally, the introduction of Sle3 onto the LDLr−/− background is sufficient to mediate Th2 antibody bias, as previously seen in the triple congenic mouse model. Interestingly, our results suggest that in autoimmune disease, serum cholesterol levels may be a weak indicator of atherosclerosis progression. Future studies should be designed to study how the loss of tolerance to nuclear antigens may facilitate atherosclerosis and to determine how dyslipidemia may affect the autoimmune status of other congenic lupus mouse models. These studies will ultimately advance our knowledge regarding autoimmune-mediated atherosclerosis and lead to the development of therapeutic agents designed to treat both lupus and atherosclerosis.
Flow cytometry experiments were performed in the VMC Flow Cytometry Shared Resource. The VMC Flow Cytometry Shared Resource is supported by the Vanderbilt Ingram Cancer Center (P30 CA68485) and the Vanderbilt Digestive Disease Research Center (DK058404). This research was funded by the National Institute of Health (NIH) Grant 1R01HL088364-01A1 (to A.S.M.) and the NIH 5T32 HL007411-28 Training Grant in Cardiovascular Mechanisms: Training in Investigation (to N.S.W.).