SEARCH

SEARCH BY CITATION

Keywords:

  • apolipoprotein B;
  • atherosclerosis;
  • autoimmunity;
  • regulatory T cells;
  • vaccine

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Abstract.  Wigren M, Kolbus D, Dunér P, Ljungcrantz I, Söderberg I, Björkbacka H, Fredrikson GN, Nilsson J. (Malmö University Hospital, Lund University; Malmö University, Malmo; Sweden) Evidence for a role of regulatory T cells in mediating the atheroprotective effect of apolipoprotein B peptide vaccine. J Intern Med 2011; 269: 546–556.

Objectives.  Autoimmune responses against oxidized low-density lipoprotein are considered to play an important pro-inflammatory role in atherosclerosis and to promote disease progression. T-regulatory cells (Tregs) are immunosuppressive cells that have an important part in maintaining self-tolerance and protection against autoimmunity. We investigated whether aBp210, a prototype atherosclerosis vaccine based on a peptide sequence derived from apolipoprotein B, inhibits atherosclerosis through the activation of Tregs.

Design.  Six-week-old Apoe−/− mice were immunized with aBp210 and received booster immunizations 3 and 5 weeks later, as well as 1 week before being killed at 25 weeks of age.

Results.  At 12 weeks, immunized mice had increased expression of the Treg marker CD25 on circulating CD4 cells, and concanavalin A (Con A)-induced interferon-γ, interleukin (IL)-4, and IL-10 release from splenocytes was markedly depressed. At 25 weeks, there was a fivefold expansion of splenic CD4+ CD25+ Foxp3 Tregs, a 65% decrease in Con A-induced splenic T-cell proliferation and a 37% reduction in the development of atherosclerosis in immunized mice. Administration of blocking antibodies against CD25 neutralized aBp210-induced Treg activation as well as the reduction of atherosclerosis.

Conclusions.  The present findings demonstrate that immunization of Apoe−/− mice with the apolipoprotein B peptide vaccine aBp210 is associated with activation of Tregs. Administration of antibodies against CD25 results in depletion of Tregs and blocking of the atheroprotective effect of the vaccine. Modulation in atherosclerosis-related autoimmunity by antigen-specific activation of Tregs represents a novel approach for treatment of atherosclerosis.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Arterial inflammation induced by accumulation and oxidative modification of lipoproteins has for many years been recognized as a key factor in the development of atherosclerosis [1]. More recently, it has been revealed that adaptive immunity plays an important role in modulating this inflammation and that both protective and disease-promoting immune responses exist [2–4]. Evidence from animal models and human plaque studies shows that Th1 type T cells drive arterial inflammation and contribute to lesion progression, whereas the role of Th2 type T cells is less clear [5–7]. Effector T cells colocalize with major histocompatibility complex class II-expressing macrophages and dendritic cells in atherosclerotic lesions, indicating active interactions between T cells and antigen-presenting cells [8–10]. The presence of local antigen presentation is also supported by the demonstration of clonal expansion of T cells in plaques from both humans and Apoe−/− mice [11, 12]. About 10% of T-cell clones derived from human plaques are specific for oxidized low-density lipoprotein (LDL), suggesting that these particles are key auto-antigens in atherosclerosis [13]. Transfer of T cells reactive to oxidized LDL aggravates atherosclerosis in mice [14]. However, attempts to characterize the role of oxidized LDL autoimmunity, by immunizing hypercholesterolaemic animals with oxidized LDL, unexpectedly resulted in an inhibition of atherosclerosis [15–18]. This finding demonstrates that protective immune responses also exist and focused attention on the possibility of developing immunomodulatory therapy for atherosclerosis [19, 20].

We have previously identified certain peptide sequences in apolipoprotein (apo) B-100 as major targets for autoimmune responses in humans [21] and demonstrated that prototype vaccines based on some of these peptides inhibit the development of atherosclerosis in Apoe−/− mice [22–24]. Apo B peptide immunization is generally associated with induction of a peptide-specific Th2-type immunoglobulin (Ig)G response [24]. Treatment of Apoe−/− mice with recombinant IgG demonstrating identical epitope specificity also inhibits atherosclerosis [25], suggesting that the protective effect of immunization may be mediated by the antibody response. However, recent studies in mice transgenic for human apo B have shown that immunization with apo B peptide vaccines can protect against atherosclerosis also in the absence of an antibody response [26]. In this study, we have analysed alternative pathways that may mediate the protective effect of apo B peptide vaccines, focusing on the possible role of regulatory T cells (Tregs). Tregs are immunosuppressive cells that play an important role in maintaining self-tolerance and protection against autoimmunity [27]. Depletion of Tregs aggravates atherosclerosis [28], whereas Treg transfer has the opposite effect [29]; these findings support a functional role of these cells in atherosclerosis [30]. To determine whether Tregs mediate the protective effect of apo B peptide vaccines, we administered antibodies that block the action of natural Tregs (anti-CD25).

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Mice, immunization and tissue preparation

Male apo E-deficient mice (B6.129P2-Apoetm1UncN11) were purchased from Taconic Laboratory, Ry, Denmark. Mice were allowed food and tap water ad libitum. From 10 weeks of age, the animals received a high-cholesterol diet (0.15% cholesterol, 21% fat; Lantmännen, Stockholm Sweden). Mice were given subcutaneous injections (100 μL) of the aBp210 vaccine at 6, 9 and 11 weeks of age. The aBp210 vaccine contained the apoB peptide 210 (amino acids 3136–3155) conjugated to cationized bovine serum albumin (cBSA) with aluminium hydroxide (alum; Pierce, Rockford, IL, USA) as adjuvant. Injections of phosphate-buffered saline (PBS) alone served as controls. From 6 to 11 weeks of age and at week 24, mice received weekly intraperitoneal injections of neutralizing anti-CD25 antibodies (BioLegend, San Diego, CA, USA). Treatment with rat IgG1 antibodies (BioLegend) directed against trinitrophenol served as an irrelevant antibody control. At 12 weeks of age, a venous blood sample was drawn from the saphenous vein. Mice were killed at 12 or 25 weeks of age by intraperitoneal injection of ketamine and xylazine. Spleens were harvested and stored in PBS on ice. Plasma was collected by cardiac puncture and stored at −80 °C until analysis. Mice killed at 25 weeks were whole-body perfused with PBS followed by Histochoice (Amresco, Solon, OH, USA), and the descending aorta was then dissected free of connective tissue and fat, cut longitudinally, mounted en face and stored in Histochoice [25]. The Animal Care and Use Committee of Lund University approved the experimental protocols used in this study.

Cell preparation and cultures

Splenocytes in single-cell suspension were prepared by pressing the spleen through a 70-μm cell strainer (BD Falcon, Franklin Lakes, NJ, USA). Erythrocytes were removed using red blood cell lysing buffer (Sigma, St. Louis, MO, USA). Cells were cultured in culture medium (RPMI 1640 medium containing 10% heat-inactivated foetal calf serum, 1 mmol L−1 sodium pyruvate, 10 mmol L−1 Hepes, 50 U penicillin, 50 μg mL−1 streptomycin, 0.05 mmol L−1β-mercaptoethanol and 2 mmol L−1 L-glutamine; GIBCO, Paisley, UK) in 96-well, round-bottom plates (Sarstedt, Nümbrecht, Germany). For proliferation assays, splenocytes (2 × 105 cells/well) were cultured alone, with 2.5 μg mL−1 concanavalin A (Con A; Sigma) or with CD3/CD28 beads (Invitrogen, Paisley, UK) at a cell-to-bead ratio of 1 : 1 for 90 h. To measure DNA synthesis, the cells were pulsed with 1 μCi [methyl-3H] thymidine (Amersham, Piscataway, NJ, USA) during the last 16 h of culture; macromolecular material was then harvested on glass fibre filters using a FilterMate harvester (Perkin Elmer, Buckinghamshire, UK) and analysed using a liquid scintillation counter (PerkinElmer,). In parallel, 3 × 105 splenocytes/well were cultured alone or with 2.5 μg mL−1 Con A for 72 h, and cytokine concentrations were measured in the cell culture supernatant using a Th1/Th2 9-plex (Meso Scale Discovery, Gaithersburg, MD, USA) according to the manufacturer’s instructions. The lower detection limit for all cytokines in this assay was about 0.5 pg mL−1.

Flow cytometry

Blood leucocytes or splenocytes were stained with fluorochrome-conjugated antibodies and analysed with a CyAn ADP flow cytometer (Beckman Coulter, High Wycombe, UK). We used the following antibodies in these experiments: Pacific Blue-conjugated anti-CD4, allophycocyanin-conjugated anti-CD25, phycoerythrin/Cy7-conjugated anti-CD4, phycoerythrin/Cy5- conjugated anti-ICOS (all from BioLegend) and Pacific Blue-conjugated anti-Foxp3 (eBioscience, San Diego, CA, USA).

Staining of the descending aorta

En face preparations of the descending aorta were washed in distilled water, dipped in 78% methanol and stained for 40 min in 0.16% Oil Red O dissolved in 78% methanol/0.2 mol L−1 NaOH. The stained plaque areas were quantified blindly using Image pro plus 4.5 software (Media Cybernetics, Bethesda, MD, USA).

Immunohistochemistry of tissue sections

Sections (thickness 10 μm) were cut from the aortic root for immunohistochemical staining of atherosclerotic plaque. Each series of experiments included up to six sections per mouse. The sections were refixed in Histochoice for 10 min, washed in dH2O and dipped in 60% isopropanol. Slides stained with rat anti-mouse MOMA-2 antibodies (monocyte/macrophage; BMA Biomedicals, Augst, Switzerland) were first fixed in ice-cold acetone for 10 min, washed in PBS for 5 min and then blocked with 10% mouse serum in PBS for 30 min and quickly dipped in PBS. Biotinylated rabbit anti-rat IgG (Vector Laboratories, Burlingame, CA, USA) was used as a secondary antibody with DAB detection kit for colour development (Vector). The primary or secondary antibodies were omitted as controls. Immunostained areas were quantified with BioPix iQ 2.0 software (Biopix AB, Gothenburg, Sweden).

Plasma cholesterol and triglyceride

Total plasma cholesterol and triglyceride levels were quantified with colorimetric assays, Infinity™ Cholesterol and Triglyceride (INT), respectively (Thermo Scientific, Waltham, MA, USA).

Statistics

Data are presented as mean ± standard deviation. Student’s two-tailed t-test was used for normally distributed samples and the Mann–Whitney rank sum test for skewed data. Statistical significance was considered at the level P ≤ 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

The development of atherosclerosis vaccines based on native and aldehyde-modified apo B-100 peptides has previously been described [22–24, 26]. The vaccine used in this study (aBp210) consisted of the human apo B-100 amino acid sequence 3136–3155 conjugated to cBSA with alum as adjuvant. Apoe−/− mice were first immunized at 6 weeks of age, and booster injections were given 3 and 5 weeks later. High-fat diet was provided from 1 week after the second booster immunization to allow an immune response to become activated before the induction of a more-severe hypercholesterolaemic state. Because recent studies have shown that alum may capture apo B peptides derived from oxidized LDL at the injection site, we could not use alum as a control in the present mechanistic studies [31]. Accordingly, control mice were injected with PBS alone. To determine the role of CD25+ T cells in mediating the protective effects of the vaccine, we gave weekly intraperitoneal injections of 100 μg neutralizing CD25 antibody between 6 and 11 weeks of age as well as 1 week before the animals were killed. To analyse the effect of aBp210 on early Treg activation, we measured the presence of CD4+ CD25 +  T cells in the circulation 1 week after the last booster immunization. The fraction of CD4 +  T cells expressing CD25 was found to be significantly increased in aBp210-immunized mice, compared with controls (11.7 ± 1.2% versus 7.5 ± 1.7%, P < 0.0001; Fig. 1). This increase was completely abolished in mice that received anti-CD25 IgG (Fig. 1). In nonimmunized mice, the fraction of CD4 +  cells expressing CD25 was lower in mice treated with anti-CD25 IgG than in controls (5.4 ± 1.7% versus 7.5 ± 1.7%, P < 0.05; Fig. 1), confirming previous findings that these antibodies act by depletion of Tregs [28]. To further characterize how aBp210 influenced immune activation in Apoe−/− mice, we injected a separate set of animals with aBp210 (n = 12) or PBS (n = 12) and determined the expression of interferon (INF)-γ (Th1), interleukin (IL)-4, IL-5 (Th2) and IL-10 (Th2/Th3) by cultured splenocytes and in plasma. The mice were killed 1 week after the second booster immunization. Treatment with aBp210 almost completely inhibited Con A-induced expression of IFN-γ and IL-10 in cultured splenocytes and markedly reduced Con A-induced release of IL-4 (Table 1). Basal secretion of INF-γ and IL-10 was also lower in splenocytes from aBp210-treated mice. The plasma level of IL-5 was higher in aBp210-treated mice, whereas no differences were observed for INF-γ, IL-4 and IL-10.

image

Figure 1. Effect of aBp210 and CD25 blocking antibodies on blood leucocytes in 12-week-old Apoe−/−mice. Blood leucocytes from mice treated with aBp210, CD25 or control antibodies or PBS were analysed with flow cytometry to determine the percentage of CD4+ blood cells expressing CD25. Representative dot plots show cells from PBS-treated and aBp210-treated mice, and from mice treated with aBp210 plus anti-CD25 antibodies. The percentages given are relative to the total CD4 T-cell population.

Download figure to PowerPoint

Table 1. Cytokines in media from spleen T cells cultured with or without Con A and in plasma from 12-week-old apoe−/− mice
 PBSaBp210
BasalCon APlasmaBasalCon APlasma
  1. *P < 0.05, **P < 0.01, ***P < 0.001.

IFNg (pg mL−1)3.80 ± 4.691127.38 ± 390.351.19 ± 0.801.22 ± 2.30*91.38 ± 107.06***2.53 ± 1.71
IL-10 (pg mL−1)23.27 ± 18.85189.33 ± 104.0883.77 ± 16.081.66 ± 4.68***15.75 ± 16.26***101.27 ± 48.60
IL-4 (pg mL−1)3.10 ± 3.4316.57 ± 6.441.75 ± 1.511.73 ± 2.757.2 ± 2.06**4.99 ± 3.12
IL-5 (pg mL−1)0.11 ± 0.217.40 ± 9.6811.04 ± 4.456.43 ± 10.979.22 ± 15.1918.07 ± 6.65*

The effect of aBp210 on atherosclerosis was assessed by Oil Red O staining of the descending aorta in mice killed at 25 weeks of age. The mice received a final injection of aBp210 with or without CD25 antibodies 1 week before they were killed to facilitate detection of treatment-specific immune responses. Treatment with aBp210 reduced the development of atherosclerosis in the aorta by 37% (P < 0.05) compared with PBS-treated controls (Fig. 2a). The protective effect of aBp210 immunization was completely reversed by administration of CD25 blocking antibodies. Treatment with CD25 blocking antibodies per se has previously been reported to increase atherosclerosis in Apoe−/− mice [28]. To determine whether the increased atherosclerosis in the aBp210/CD25 IgG group was owing to a general increase in atherosclerosis rather than to an inhibition of the protective effect of immunization, we injected nonimmunized mice with CD25 IgG. Treatment with CD25 IgG resulted in a nonsignificant 18% increase in atherosclerosis in nonimmunized mice (anti-CD25 versus PBS group) compared with the 80% increase observed in aBp210-immunized mice (aBp210/anti-CD25 versus aBp210, P < 0.05; Fig. 2a). We measured macrophage (MOMA-2) immunoreactivity in subvalvular lesions to assess the effect of treatment on plaque inflammation. There was no difference in MOMA-2 immunoreactivity between aBp210-immunized mice and nonimmunized controls. However, plaque inflammation was markedly increased in mice given CD25 antibodies (49.8 ± 12.4% versus 21.8 ± 10.0% MOMA-2-stained plaque area in nonimmunized PBS controls, P < 0.0005, Fig. 2b). This pro-inflammatory effect of CD25 antibody treatment was almost completely inhibited by concomitant aBp210 immunization. There was no significant difference in subvalvular plaque area between any of the groups (data not shown).

image

Figure 2. Effect of aBp210 and CD25 antibodies on atherosclerosis and plaque inflammation in Apoe−/− mice. (a) Plaque areas in descending aortas from 25-week-old Apoe −/− mice treated with PBS, aBp210 and/or blocking CD25 antibodies were assessed by en face Oil Red O staining, and the percentage stained area of the total aortic area was determined by computerized image analysis. (b). Subvalvular lesions from 25-week-old Apoe −/− mice treated with PBS, aBp210 and/or blocking CD25 antibodies were stained with MOMA-2 antibody, and the stained areas were quantified by computerized image analysis.

Download figure to PowerPoint

To further characterize the effect of aBp210 treatment on Tregs, we analysed the expression of CD25 and FoxP3 in spleen CD4 +  T cells from 25-week-old mice. Immunization with aBp210 was associated with a several fold increase in CD4+ CD25+ Foxp3 +  natural Tregs in the spleen (Fig. 3a). This increase was fully inhibited by administration of CD25 blocking antibodies. A correlation analysis that included all animals in the PBS control and aBp210-immunized groups (with or without antibodies) revealed a trend towards an inverse association between spleen Tregs and atherosclerosis (r = 0.31, P = 0.058; Fig. 3b). Immunization with aBp210 was also associated with a modest reduction in ICOS expression on spleen CD4 +  T cells (28.5 ± 8.0% versus 36.8 ± 6.8% in PBS control mice, P < 0.05). Analysis of ICOS expression demonstrated that aBp210 immunization induced ICOS expression on CD25 +  T cells (23.3 ± 7.7% versus 12.2 ± 2.5% in PBS control mice, P = 0.0005), whereas it markedly downregulated ICOS expression on CD25 T cells (7.7 ± 1.6% versus 35.8 ± 4.7% in PBS control mice, P < 0.0001). Administration of anti-CD25 IgG resulted in a marked reduction in CD4+ T-cell ICOS expression (8.4 ± 2.6% versus 36.8 ± 6.8% in PBS control mice, P < 0.0001).

image

Figure 3. Effect of aBp210 and CD25 antibodies on spleen T cells in 25-week-old Apoe−/−mice and correlation with atherosclerosis. (a) Splenocytes from mice treated with PBS, aBp210 and/or blocking antibodies were analysed with flow cytometry to determine the percentage of spleen CD4+ T cells expressing CD25+ Foxp3 + . Representative dot plots show cells from PBS-treated (left) and aBp210-treated (right) mice. The percentages given are relative to the total CD4 T-cell population. A bi-exponential transformed scale is used to show the negative decade. (b) The correlation between spleen CD4+ T cells expressing CD25+ Foxp3+ and en face Oil Red O staining of the descending aorta is shown. Data from all animals in the PBS control and aBp210-immunized groups, with or without antibodies, are presented.

Download figure to PowerPoint

The possible suppressive action of aBp210-mediated Treg activation was studied by analysing the effect of Con A and anti-CD3/CD28-induced T-cell proliferation. Con A-induced T-cell proliferation was reduced by 66% in aBp210-immunized mice compared with PBS-treated controls (proliferative index 4.4 ± 0.5 versus 13.0 ± 3.1, P < 0.01, Fig. 4a). The inhibitory effect of aBp210 on T-cell proliferation was not affected by anti-CD25 IgG. Similar observations were made when T cells were stimulated with anti-CD3/CD28 IgG instead of Con A (Fig. 4b).

image

Figure 4. Effect of aBp210 and CD25 antibodies on Con A- and CD3/CD28-induced splenocyte proliferation. Cultured splenocytes from 25-week-old Apoe−/− mice, treated with PBS, aBp210 and/or neutralizing CD25 antibodies were stimulated with Con A (a) or CD3/CD28 beads (b). T-cell proliferation index is expressed as thymidine incorporation ratio between stimulated and nonstimulated cells.

Download figure to PowerPoint

Splenocytes isolated from aBp210-treated mice released increased amounts of IL-10 when exposed to Con A in culture (Fig. 5), whereas no such increase was observed when the mice were also given blocking antibodies against CD25. IL-5 was also increased following Con A stimulation of splenocytes from aBp210-treated mice, but this was not affected by administration of CD25 antibodies. The release of IFN-γ and IL-4 was not changed in response to aBp210 immunization, but decreased levels of IFN-γ were seen in medium from mice also given CD25 antibodies (Fig. 5).

image

Figure 5. Effect of aBp210 and CD25 antibodies on cytokine levels in cell culture medium from cultured spleen T cells. Quantification of IFN-γ (a), IL-10 (b), IL-4 (c) and IL-5 (d) in cell culture medium from Con A-stimulated cultured spleen T cells from 25-week-old Apoe−/− mice treated with PBS, aBp210 and/or blocking antibodies. Cytokine levels were analysed by multiplex technology.

Download figure to PowerPoint

Mice immunized with aBp210 had higher plasma levels of IL-5 than controls, whereas no differences were observed in the levels of IFN-γ, IL-10 or IL-4. It is interesting that increased plasma levels of all four cytokines were observed when aBp210-treated mice were injected with anti-CD25 blocking antibodies (Table 2), whereas no corresponding increase was seen in Con A-stimulated splenocytes from the same animals (Fig. 5). Cholesterol levels were higher in immunized mice than in PBS-treated controls (Table 2).

Table 2. Plasma cholesterol, triglycerides and cytokines in 25-week-old apoe−/− mice
 PBSaBp210aBp210 + anti-CD25
  1. *P < 0.05, **P < 0.01.

Cholesterol (mmol L−1)18.47 ± 2.7021.39 ± 3.39*24.67 ± 3.73*
Triglycerides (mmol L−1)0.57 ± 0.120.70 ± 0.180.66 ± 0.08
IFNy (pg mL−1)4.17 ± 6.511.73 ± 1.4790.44 ± 78.63**
IL-10 (pg mL−1)87.94 ± 155.5546.08 ± 84.87876.02 ± 771.07**
IL-4 (pg mL−1)2.26 ± 5.410.33 ± 0.6071.97 ± 59.63**
IL-5(pg mL−1)7.45 ± 5.4318.2 ± 13.79*55.45 ± 39.36*

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Targeted immunotherapy represents a novel approach for the treatment of atherosclerosis [19, 20]. Vaccines containing peptide sequences derived from the LDL protein apo B-100 are one example of such therapies that are presently being developed for clinical testing. A limiting factor in the clinical development of these novel therapies has been the incomplete understanding of their mechanism of action. This is of critical importance not only for the development of optimal vaccine formulations but also for the characterization of biomarkers required to monitor the response to treatment. Here, we present evidence that the effect of aBp210, a prototype apo B peptide–based atherosclerosis vaccine, is mediated through the activation of Tregs. These cells constitute a family of CD25- and Foxp3-expressing immunosuppressive T cells that have an important role in limiting autoimmunity [27, 32]. Six weeks after the first immunization, an increased expression of CD25 was found on circulating CD4+ T cells, and the responsiveness of splenocytes to Con A was almost completely inhibited. In animals killed at 25 weeks of age, CD4+ CD25+ Foxp3+ Tregs constituted 13.5% of the CD4+ spleen cell population in aBp210-treated mice but only 1.7% in control mice. This was associated with an increased release of the anti-inflammatory cytokine IL-10, a reduced T-cell expression of the activation marker ICOS, and a reduced proliferative response to Con A and anti-CD3/CD28. Moreover, analysis of ICOS expression of CD25 +  and CD25 T cells suggested that immunization with aBp210 selectively activates Tregs whilst downregulating CD25 effector T cells. Taken together, these findings demonstrate that aBp210 treatment results in an activation of Tregs and an associated inhibition of T-cell responsiveness. The inhibition of CD25 effector T cells was clearly much larger than could be accounted for by an effect on p210-specific T cells alone. Accordingly, inhibition of non-p210-specific T cells must also occur via a bystander mechanism.

To test the functional role of Tregs in mediating the protective effect of aBp210 immunization, we administered CD25 blocking antibodies. In the present studies, aBp210 treatment reduced the development of atherosclerosis in the aorta by 37%. Administration of CD25 antibodies resulted in depletion of the CD4+ CD25+ Foxp3 +  Treg population and a complete inhibition of the protective effect of aBp210 on plaque development in the aorta. These observations imply that Tregs play a critical role in mediating the effect of the vaccine. In view of their important regulatory role in protection against autoimmunity, Tregs represent a logical target for immunotherapy in atherosclerosis. There is solid evidence that Th1-driven immune responses against altered self antigens generated by hypercholesterolaemia play a key role in the development of atherosclerosis [6, 7, 33, 34]. Mallat et al. [35] first suggested that atherosclerosis may be promoted by an imbalance between regulatory and pathogenic immunity. This hypothesis has gained support as a result of several studies demonstrating an atheroprotective role of Tregs. Deficiency of the costimulatory molecules required for Treg generation and homeostasis (CD28, CD80, CD86 and ICOS) results in a more aggressive development of atherosclerosis in hypercholesterolaemic mice [28, 36]. Increased development of atherosclerosis has also been observed following depletion of Tregs by anti-CD25 treatment [28]. By contrast, induction of an antigen-specific (ovalbumin) Treg response was found to significantly reduce atherosclerosis in Apoe−/− mice [29]. Inhibition of atherosclerosis by oral immunization with HSP60, [37–39] as well as with oxidized LDL, [40] has also been associated with an activation of Tregs. The functional suppressive properties of Tregs from Apoe−/− mice are reduced compared with those from wild-type mice [41]. In addition, the proportion of Tregs is much lower in human atherosclerotic plaques than in normal tissue or other inflammatory lesions [42].

Because previous studies have shown that depletion of Tregs by anti-CD25 treatment results in a more aggressive development of atherosclerosis, [28] we investigated the possibility that the effect of anti-CD25 in our studies was attributed to a general aggravation of the disease process rather than to a neutralization of the protective effect of aBp210 treatment. We observed a modest 18% increase in atherosclerosis in nonimmunized mice treated with anti-CD25, and the total plaque area was similar to that found in aBp210/anti-CD25-treated mice. This indicates that the increased development of atherosclerosis in aBp210/anti-CD25-treated animals could be explained by a combination of neutralization of the protective immunization and a modest increase in the activity of the underlying disease process. Treatment with CD25 IgG was associated with a several fold increase in the plasma levels of IFN-γ, IL-4, IL-5 and IL-10, suggesting a loss of suppression of both Th1 and Th2 immunity. We did not observe a corresponding increase in cytokine release in Con A-stimulated splenocytes from the same animals. However, it is likely that any possible enhanced basal cytokine secretion by these cells may have been masked by the effects of stimulation with Con A.

The suppressive function of CD4+ CD25+ Foxp3 +  natural Tregs seems to require only cell–cell contact or proximity in vitro, whereas the in vivo function of these cells depends on secretion of IL-10 and tumour growth factor-β [43]. Immunization with aBp210 both enhanced Con A-induced secretion of IL-10 from splenocytes and inhibited Con A- and CD3/CD28 IgG-induced proliferation of splenocytes in the present experiments. Treatment with CD25 antibodies inhibited the increase in IL-10 release but did not restore the proliferative capacity of spleen cells. These results imply that the atheroprotection of aBp210 immunization may primarily involve an increased secretion of IL-10 by Tregs. However, as CD25 antibodies have been shown to inhibit the response of CD4 T cells to Con A in vitro [44] (presumably by blocking the IL-2 receptor), we cannot excluded the possibility that the reduced responsiveness of cultured splenocytes from aBp210/anti-CD25-treated mice is explained by CD25 antibodies remaining bound to the surface of T-effector cells even after isolation. Another possibility is that CD25 antibody treatment has deleted not only CD25-expressing Tregs but also a minor population of activated CD25-expressing effector cells with the capacity to proliferate and secrete INF-γ.

There are several limitations in the present study that need to be considered. Most importantly, it is not clear whether Tregs play the same atheroprotective role in humans as in mice. Tregs have been reported to be reduced in patients with acute coronary syndrome, [45] but there are no studies relating Tregs to risk for future development of cardiovascular events in humans. Accordingly, it remains to be determined whether the activation of Tregs is also associated with protection against atherosclerosis in humans. Moreover, because our antibody blocking approach targeted the CD25 molecule, we cannot completely exclude the possibility that the protective effect of aBp210 immunization is mediated by CD25-expressing effector T cells rather than by Tregs. However, available evidence suggests that the atheroprotective effect of immunization is not mediated through the activation of Th1 cells because ablation of different factors involved in Th1 activation has resulted uniformly in decreased atherosclerosis [3]. Although the observation of reduced development of atherosclerosis in mice deficient in the Th2 cytokine IL-4 has suggested a pro-atherogenic role of Th2 immunity, [46] we cannot exclude the possibility that activation of CD25 on Th2 cells may be involved in mediating the protective response to aBp210. This possibility is to some extent supported by the increased release of IL-5 from Con A-stimulated splenocytes and increased plasma levels of IL-5 observed in aBp210-immunized mice. However, as treatment with CD25 blocking antibodies did not inhibit aBp210-induced IL-5 expression, it appears unlikely that CD25 was involved in mediating activation of Th2 immune responses in the present study. In this context, it is interesting to note that immunization with aBp210 inhibited the increase in plaque inflammatory activity caused by CD25 antibody treatment. This effect is unlikely to be mediated by Tregs. One possibility is that this effect of aBp210 involves activation of Th2 immunity. Binder et al. [47] have reported that the atheroprotective effect of immunization with malondialdehyde-modified LDL is dependent on IL-5. Immunization of mice with apo B peptides is associated with generation of peptide-specific Th2-type IgG [24], and treatment of mice with recombinant IgG of the same specificity has been shown to reduce inflammation in subvalvular lesions [25]. Another limitation of the present study that needs to be considered is that alum could not be used as a control for specificity. The aBp210 vaccine consists of the apo B 3136–3155 amino acid sequence conjugated to cBSA and the adjuvant alum. Several studies have demonstrated that apo B peptide–based vaccines reduce atherosclerosis by 30–60% compared with alum alone [22–24, 26]; this clearly shows the peptide-dependent specificity of the effect. However, alum has also been shown to have atheroprotective properties in itself, and Khallou-Laschet et al. [48] found that injection of alum alone significantly decreased atherosclerosis compared with control saline injections. We have recently explored the mechanisms through which alum inhibits atherosclerosis and demonstrated that injection of alum is associated with oxidation of LDL and capture of oxidized LDL antigens (including modified apo B peptides) in the subcutaneous alum precipitate. This was subsequently associated with a trend towards increased plasma levels of p210-specific IgG1, as well as with an expansion of the CD4+ CD25+ Foxp3 +  natural Treg population in the spleen. It is interesting that no such effects were observed when wild-type C57BL/6 mice were injected with alum. Collectively, these observations imply that alum in a hypercholesterolaemic environment may mimic some of the effects of aBp210, and consequently that alum cannot be used as an antigen-specificity control in studies characterizing the mechanism of action of apo B peptide–based vaccines. Plasma cholesterol levels were higher in immunized mice than in PBS-treated controls, and the highest levels were observed in immunized mice given CD25 blocking antibodies. We cannot rule out the possibility that this in part may have contributed to the increased development of atherosclerosis in the latter group compared with mice given aBp210 alone.

In the present study, immunization with aBp210 reduced plaque size in the aorta, whilst the size of subvalvular plaques was unchanged. This observation is consistent with the results of several other studies of oxidized LDL and apo B peptide immunization, which demonstrated the variable effects of immunization at different sites of the arterial tree [19]. The possibility that various antigens may have site-specific effects remains to be fully explored.

In conclusion, our results show that immunization of Apoe−/− mice with the apo B peptide vaccine aBp210 is associated with activation of Tregs. Co-administration of antibodies against CD25 results in depletion of Tregs and blocking of the atheroprotective effect of the vaccine. Modulation in atherosclerosis-related autoimmunity by antigen-specific activation of Tregs represents a novel approach for the treatment of atherosclerosis. The present findings also imply that Tregs represent possible biomarkers for cardiovascular protection and that they may be used to monitor the effect of a possible future atherosclerosis vaccine.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

This study was supported by grants from the Swedish Medical Research Council, the Swedish Heart-Lung Foundation, the Crafoord Foundation, VINNOVA, the Knut and Alice Wallenberg Foundation, the Bergvall Foundation, the Swedish Society of Medicine, the Royal Physiographic Society, Immunath (the European Community’s Sixth Framework Programme), the Albert Påhlsson Foundation, the Malmö University Hospital Foundation, the Lundström Foundation and Cardiovax.

Disclosures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Jan Nilsson appears as co-inventor on patents for the use of apo B peptide vaccines in atherosclerosis.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References