Immunization of Chlamydia pneumoniae (Cpn)-Infected Apobtm2SgyLdlrtm1Her/J Mice with a Combined Peptide of Cpn Significantly Reduces Atherosclerotic Lesion Development

Objective To investigate the antigenic effect of a peptide containing two epitopes of Chlamydia pneumoniae (Cpn) on atherosclerotic lesion formation in mice infected with Cpn. Materials and Methods Six-week-old Apobtm2SgyLdlrtm1Her/J mice were immunized using a repetitive immunization multiple-sites strategy with KLH-conjugated peptides derived from the major outer membrane protein and the putative outer membrane protein 5 of Cpn. Mice were fed a high-fat diet and infected with Cpn twice during the 10-week diet period. Lesions were evaluated histologically; local and systemic immune responses were analyzed by immunohistochemistry of aorta samples and cytokine measurements in plasma samples and splenocyte supernatants. Results Mice immunized with the combined Cpn peptide showed a greater reduction in lesion size compared to mice immunized with either epitope alone [54.7% vs 39.8% or 41.72%] and was also associated with a significant decrease in lesion area in descending aortas compared with those in controls (88.9% for combined Cpn peptide, 81.9% for MOMP peptide and 75.7% for Omp5, respectively). This effect was associated with a shift in the cellular composition of plaques towards decreased inflammatory cell and increased regulatory T-cell content. Additionally, the effect was also connected with decreased secretion of proinflammatory cytokines and increased production of anti-inflammatory cytokines demonstrated in plasma and in supernatant on stimulated spleen cells. Conclusions Atherosclerotic lesion formation may be promoted by Cpn infection in the presence of a high-fat diet, and reduced by immunization with the combined Cpn peptide. The combined peptide has more potential than either epitope alone in reducing atherosclerotic lesion development through Treg expansion.


Introduction
Chlamydia pneumoniae (Cpn) [1] is an important human pathogen that causes atypical pneumonia and is associated with various chronic inflammatory diseases such as atherosclerosis, a major cause of cardiovascular disease and death in the Western world [2][3][4][5][6]. Although the epidemiological importance of Chlamydia infection in atherosclerosis is not well defined, the potential role of Cpn in coronary atherosclerosis may be related more to acceleration of the disease or to the systemic effects of persistent infection than to sudden initiation of infarction by acute infection [7]. However, the theoretical role of Cpn in acceleration of atherosclerosis is still controversial [8][9][10]. Although an association between Cpn infection and coronary atherosclerosis has been reported, the association is less clear for the effect of peptide antigen derived from Cpn on the formation of atherosclerotic lesion. In addition, an epitope of the major outer membrane protein (MOMP) of Cpn (AA 67-74: GDYVFDRI) and the putative outer membrane protein 5 (Omp5) of Cpn (AA 284-292: QAVANGGAI) share high homology, with two sequence locations of ApoB protein (http://web.expasy.org/sim/). ApoB protein plays a crucial role in atherosclerosis as immunization with some peptides derived from ApoB protein reduce atherosclerotic lesion in several mouse models. Indeed, this molecular mimicry (share high homology) was recently demonstrated in our laboratory in which an epitope containing both sequences of AA 67-74 (GDYVFDRI) and AA 284-292 (QAVANGGAI) has an effect on atherosclerotic lesion reduction in a protein scaffold in non-infected mice with Chlamydophia [11]. In this study, we investigate the effect of a linear peptide containing these two putative epitopes derived from MOMP and Omp5 of Cpn on atherosclerotic lesion formation in Cpn-infected Apob tm2Sgy Ldlr tm1Her /J mice.

Materials and Methods
The immunizing peptides derived from MOMP (AA 67-74: GDYVFDRI, designated as MOMP peptide) and Omp5 (AA 283-291: QAVANGGAI, designated as Omp5 peptide), and a combined peptide containing the MOMP and the Omp5 peptides (designated as a combined Cpn peptide) coupled by a polyglycine [(Gly)5] linker, were used in this study in a Keyhole limpet hemocyanin (KLH)-conjugated form. All of the peptides used in the study, including ApoB peptide and human HSP60 (hHSP60) peptide, were synthesized by Severn biotech Ltd (Worcestershire, UK).

Animal Experiments
The experiments were approved by the Animal Welfare Committee of the University of Szeged and conform to the Directive 2010/63/EU of the European Parliament.
Apob tm2Sgy Ldlr tm1Her /J mice (these mice produce ApoB100 only, and are LDL receptor deficiency).were used in our study in a total of 5 groups (3 sample and 2 control groups). Each group included 6 mice (5-6-week-old males; similar body weight, 32.2662.12 g [measured at the end of the experiment]) and the experiment was repeated. Mice were immunized with KLHconjugated peptides mixed with Alum adjuvant subcutaneously according to a repetitive immunization multiple sites strategy (RIMMS) as described earlier [11,12]. For infection, mice were inoculated intranasally with 2610 6 inclusion forming units (IFU) of Cpn (CV-6, cardiovascular strain) [13] in 25 mL of phosphate buffered saline (PBS) at week 4 and at week 8. This dose was chosen based on survival and symptoms observed in mice after infection with different infection doses (Table S1). CV-6 strain of Cpn was propagated in HEp-2 cells and partially purified as described earlier [14]. The mice were sacrificed at the end of week 12 (a high-fat diet was started at week 2 and continued for 10 weeks). For detection of Cpn-specific DNA, polymerase chain reaction (PCR) was performed as described by Tong and Sillis [15]. The MOMP of Cpn was chosen as a target for amplification in a nested PCR. All primers were synthesized in Life Technologies Ltd (Paisley, UK). The external primers (Table S2) amplified a 333 base-pair product (first-stage PCR) from the genomic DNA purified from lung homogenates of tested mice in both infected and non-infected (negative control) groups. The internal primers amplified a 207 base-pair product (second-stage PCR) using the first-stage PCR product as a template. To confirm if the mice were infected, further detection of Cpn-specific IgG in the mouse sera by indirect immunofluorescence was carried out. A HEp-2 cell (ATCC) monolayer grown on 13-mm coverslips in 24-well plates was infected with Cpn (CWL029) at a multiplicity of infection (m.o.i.) of 1. At 48 h of infection, cells were fixed with ice-cold acetone (10 min, 220uC). PBS-rehydrated cells were stained with Cpn MOMP-specific mouse monoclonal antibody (DAKO, Budapest, Hungary) and FITC-labelled anti-mouse IgG secondary antibody (SIGMA, Dorset, UK). Coverslips were examined under an ultraviolet-microscope, and evenly distributed bright fluorescing inclusions were visible. Mouse sera diluted in PBS were used as the primary antibody for staining the similarly treated monolayers and the same FITC-conjugated secondary antibody was applied to detect serum IgG binding to inclusions. Reciprocal of serum dilution producing clearly discernable inclusions was determined as Cpn-specific titre of the mouse serum. The repetitive immunization multiple-sites strategy (RIMMS) was adopted [9,10] and mice were sacrificed at the end of week 12 (a high-fat diet was started at week 2 and continued for 10 weeks).

Tissue Preparation and Antibody Response Measurements
Twelve weeks after the first immunization, tissues were harvested and mounted in Optimal Cutting Temperature (OCT) for immunohistochemical analyses and in paraffin for lesion measurement. Atherosclerosis in aortic roots was examined by Image-Pro Plus TM software, version 4.0 ((Media Cybernetics, Bethesda, MD, USA). Peptide-specific antibody titers were measured by ELISA following the manufacturer's instructions. To assess whether immunization with the combined Cpn peptide led to changes in lipoprotein profiles, the cholesterol, triglyceride, HDL-and LDL-cholesterol profiles of the mice were analyzed using a pooled plasma sample from each group after being fed a high-fat diet for 10 weeks.

Immunohistochemical Analyses, Morphometric Analyses and Quantitative Measurements of Atherosclerosis
OCT-embedded samples were used for immunohistochemical analyses. Sections of paraffin-embedded tissues were stained with hematoxylin and eosin (HE) and elastin/van Gieson (Sigma) for histological examination and were evaluated using an Olympus U-ULH Optical microscope (Olympus Optical Co. Ltd, Tokyo, Japan).
Both the atherosclerotic lesions in the aortic sinus and the lesions in the descending aortas were measured. The descending aortas were evaluated for the extent of atherosclerosis.

Measurement of Cytokines
Plasma levels of interleukin (IL)-10, transforming growth factor (TGF)-b, tumor necrosis factor (TNF)-a and interferon (IFN)-c were measured by ELISA following the manufacturer's instructions (R&D systems, Abingdon, UK). IL-10 and TNF-a levels in the lesions were quantified by immunohistochemical analyses (rat anti-mouse TNF-a and IL-10 purchased from BioLegend, CA, USA). Levels of concanavalin-A (Con A)-induced IL-10, TGF-b, TNF-a and IFN-c in splenocyte cultures were also measured.

Fluorescence Activated Cell Sorting Analysis
Spleen cells from mice immunized with either Cpn peptides or KLH (control) were used for T regular (Treg) cells measurement using a Treg detection kit (Miltenyi Biotec, Surrey, UK). Cells were analyzed with a Beckman-Coulter FC-500 Analyzer (Bachman Coulter, High Wycombe, UK).

Statistical Analyses
Data are reported as mean6standard error of the mean (6SEM), unless otherwise indicated. Figures were plotted using graph-pad Prism 5.01 and Sigma plot 9.0. For atherosclerotic lesion size, data were compared and intergroup differences were conducted using one-way ANOVA for multiple comparisons and post hoc bonferroni test. Other data were analyzed using Student's t-test (2-tailed analyses). Non-parametric distributions were analysized using Mann-Whitney U test for pairwise comparisons and the Kruskal-Wallis test for multiple comparisons. Differences between groups were considered significant at P values below 0.05.

Detection of Cpn Infection
The MOMP of Cpn was chosen as a target for amplification in nested PCR DNA from infected samples only, and produced a band of the expected size after the first PCR amplification. Similarly, in the second amplification, a positive band of the correct size was obtained only from using the first PCR product as a template originally from infected and non-infected mice ( Figure 1). Cpn DNA was detected from the lung homogenates by PCR as shown in Figure 1a, and the primers used for PCR as shown in Table S1. In addition, a genus specific epitope within Chlamydia lipopolysaccharde (LPS) was detected from the lesion sites of aortic sinus in Cpn-infected mice by Chlamydia LPS antibody (MCA 2718, AbD serotec) compared that of mice without Cpn infection (control), suggesting that Cpn bacteria were captured at the lesion site when mice were infected with Cpn bacteria as shown in Figure 1b.

Peptide-Specific Immunoglobulin G in Sera of Immunized Mice
A peptide-induced specific antibody response was observed when the MOMP peptide, the Omp5 peptide and the combined Cpn peptide were used as ELISA antigens (Figures 2A-D), when compared with a KLH control, which produced little immune response ( Figures 2C and D). The antibody immune response at week 12 was slightly lower than that at week 2 ( Figure 2A-D) apart from combined Cpn peptide-induced antibody at week 12. High antibody concentrations against either the Omp5 peptide or the MOMP peptide at week 2 were detected in immunized mice despite a relatively lower immune response in the MOMPpeptide-immunized mice than in the Omp5-peptide-immunized mice at week 12.

Peptide-Specific Immunoglobulin Subtypes G1 and G2a in Antiserum of Immunized Mice
A peptide-induced specific IgG1 response was observed, when compared with a KLH control ( Figure 2E). There was little or no IgG2a immune response ( Figure 2F) in pooled antiserum. Immunization with a KLH-conjugated ApoB and hHSP60 peptides was described earlier [11]. Sera from these experiments and sera of Cpn peptide-immunized mice were used for testing cross-reaction of peptide-specific antibodies. ApoB and Cpn peptides, individually, induced high levels of peptide-specific IgG in mouse sera two weeks after the first immunization compared to the control group immunized with KLH alone. Certain level of cross-reaction was observed between ApoB peptide and Cpn peptide antisera ( Figure 2G), as well as between Cpn peptide and ApoB peptide antisera ( Figure 2H) [16].

Effect of Treatment With Cpn Peptides on Plasma Lipid Levels
Slightly lower concentrations of cholesterol and LDL were observed in the infected control mice than in the non-infected controls (Table 1). In agreement with previous report by Blessing et al, there was no significant difference observed between the infected and non-infected mice [17]. In addition, there was little change in either triglyceride or HDL concentration observed in non-infected and infected control mice as well as in mice infected after immunization with the combined Cpn peptide. These levels were remained the same as those in infected mice after immunization with the combined Cpn peptide.

Reduction of Atherosclerotic Lesion Size in the Aortic Sinus
Representative sections from the aortic sinuses of mice are shown in Figure 3A. The lesion size in infected control mice was 5% larger than that in non-infected controls (32% vs. 27%), but it did not show significant. In contrast, the lesion size in mice immunized with the combined Cpn peptide was smaller than that in either the MOMP peptide-or the Omp5-peptide-immunized mice (14% vs. 19% and 14% vs. 18%, respectively) ( Figure 3B). The former pair of peptide antigens showed significant difference (P = 0.045), but this was not in case of the later pairs (P = 0.05). No significant difference in lesion size between mice immunized with either the MOMP peptide or the Omp5 peptide alone was observed.
Reduction in lesion size, expressed as a percentage of the lesion size in the aortic sinus of the control mice, was 54.7% following immunization with the combined Cpn peptide versus 39.8% with the MOMP peptide and 41.7% with the Omp5 peptide ( Figure 3C). In addition, the lesion reduction in mice immunized with the combined Cpn peptide was significantly greater (P = 0.046) than that in MOMP peptide-immunized mice ( Figure 3C).
In addition, we assessed oil red O (ORO) lipid levels in atherosclerotic lesions in the aortic sinus. Representative sections from the aortic sinuses of mice are shown in Figure S1A. The lipid level at the lesion sites in infected control mice was approximately 7-fold higher than in the MOMP-peptide-immunized mice (32.5% versus 4.6%, respectively), 6-fold higher than in the Omp5peptide-immunized mice (32,5% versus 5.2%, respectively) and 9fold higher than in the combined Cpn peptide-immunized mice (32.5% versus 3.6%, respectively) ( Figure S1B). Lipid level in mice immunized with the combined peptide was significantly lower than that in either the MOMP-peptide-immunized mice (4.6% versus 3.6%, P = 0.028) or the Omp5-peptide-immunized mice (5.2% versus 3.6%, P = 0.001), respectively. Furthermore, we examined the impact of treatment with Cpn peptides on the collagen in these lesions. The reduction of atherosclerosis in mice treated with these peptides was associated with an increased collagen content: 4-fold for the combined Cpnpeptide-immunized mice versus control mice (24.461.2% versus 5.960.5%; P,0.001), 3-fold for the MOMP peptide-immunized mice (19.461.3% versus 5.960.5%; P,0.001), and 3-fold for the Omp5 peptide-immunized mice (18.961.5% versus 5.960.5%; P,0.00) ( Figures S2A and B), respectively. Mice immunized with the combined peptides showed a significant collagen increase compared to mice immunized with either the MOMP-or the Omp5-peptide alone (P = 0.007 and P = 0.004, respectively).

Reduction of Atherosclerotic Lesion Size in Descending Aortas
Lesion size in the descending aortas of infected control mice was greater than that in non-infected control mice (32.0% vs. 17.0%, P,0.01) ( Figure 4A). Lesion size in the descending aortas in mice immunized with the combined peptide was smaller than that in either the MOMP-peptide-immunized (3.4% vs. 5.8%) or the Omp5-peptide-immunized (3.4% vs. 7.8%) mice respectively ( Figure 4B). There was, however, no significant difference in lesion size between mice immunized with either the MOMP peptide or the Omp5 peptide, or between mice immunized with either the MOMP peptide or the combined Cpn peptide; a significant difference (P,0.01) was found between mice immunized with the Omp5 peptide versus the combined Cpn peptide. Reduction in lesion size, expressed as a percentage of lesion size in descending aortas, was 88.9%, 81.9% and 75.7% from mice immunized with the combined Cpn peptide, the MOMP peptide and the Omp5 peptide, respectively, showing a similar trend to that observed in aorta sinus ( Figure 4C). The calculated percentage reduction in lesion size showed a significant difference (P = 0.006) between mice immunized with the Omp5 peptide versus the combined Cpn peptide ( Figure 4C).

Amount of Inflammatory Cells in The Atherosclerotic Lesions
The percentage of macrophage-occupied area in the lesions in mice immunized with the combined peptide was 7.8%, significantly lower than that in mice immunized with the Omp5 peptide (12.2%, P = 0.016), but not with the MOMP peptide (12.3%, P = 0.084) alone. In contrast, mice immunized with the KLH showed a 34.1% occupation ( Figures 5A and B).
The proportion of anti-CD11c + -stained lesion area was 5.260.4% in mice immunized with the combined Cpn peptide, lower than that with either the MOMP peptide (8.360.5%, P = 0.006) or the Omp5 peptide (8.460.5%, P = 0.002) alone, and the control group (32.962.8%, P,0.001) ( Figure 5C). In addition, the proportion of anti-CD11c + -stained lesion area co-localized with CD68 + was 71.264.5% for the combined peptide, 77.769.5% for the MOMP peptide and 74.667.0% for the Omp5 peptide, compared with 95.760.8% for the controls ( Figure 5D).
Forkhead box P3 (Foxp3) is a transcriptional regulator of CD4 + CD25 + regulatory T cells [18]. Thus, we chose to investigate whether the compromised suppressor function of Tregs observed in mice immunized with these peptides was associated with Foxp3 up-regulation. We found that the CD4 + -occupied area in the lesion was significantly reduced in mice immunized with these peptides (P,.0.001) compared with that of control mice immunized with the KLH only (33086268 mm 2 versus 10,60761435 mm 2 for the MOMP peptide, 27796525 mm 2 versus 10,60761435 mm 2 for the Omp5 and 17616359 mm 2 versus 10,60761435 mm 2 for the combined Cpn peptide, respectively) (Figures 5E and F). In addition, the CD4 + -occupied area in lesions in mice immunized with the combined Cpn peptide developed significantly smaller occupied areas in lesions than those of mice immunized with the MOMP peptide (P = 0.006) (Figures 5E and F). In contrast, the proportion of Foxp3 + area colocalized with CD4 + area in mice immunized with these peptides was increased compared with that in control mice immunized with the KLH alone, showing 21.4% (P = 0.001) for the combined Cpn peptide, 8.4% (P = 0.006) for the MOMP peptide, and 10.0% (P,0.001) for the Omp5 peptide, versus 1.1% for the control (Figures 5E and G). In addition, a statistically significant difference (P,0.05) was observed between the combined Cpn peptide and either the MOMP peptide or the Omp5 peptide ( Figures 5E and G).
Consistently, the flow cytometric analysis of spleen cells showed significantly increased expression of Foxp3 + in mice immunized with these peptides compared with that of the KLH control mice (P#0.001, Figures 5H and I). Similar to the observation in the lesions, higher expression of Foxp3 + was found in mice immunized with the combined Cpn peptide than those of mice immunized with either the MOMP peptide (P = 0.006) or the Omp5 peptide (P = 0.003).

Expression of Anti-Inflammatory Cytokines and Pro-Inflammatory Cytokines
To further characterize the effect of immunization with the combined Cpn peptide, we analyzed IL-10 expression in aortic  lesions of mice immunized with these peptides ( Figure 6A). The proportion of CD4 + cells expressing IL-10 was significantly higher in mice immunized with the combined Cpn peptide compared with control mice (.2-fold; 9.0% versus 3.6%, P = 0.024). Mice immunized with either the MOMP peptide or the Omp5 peptide showed similar proportions of IL-10 (6.0% and 5.6%, respectively) and showed significantly higher proportions in lesion sites compared with that in controls (P,0.05). Mice immunized with the combined Cpn peptide had a greater IL-10-positive area compared with either the MOMP peptide or the Omp5 peptide alone; but this increased level did not show statistical significance ( Figure 6B). Immunohistochemical analyses of TNF-a showed significantly smaller TNF-a-occupied areas in lesions of mice immunized with the combined Cpn peptide compared with controls (6.0% for the combined peptide vs. 9.2% for the MOMP peptide, 9.0% for the Omp5 peptide and 22.4% for the controls). The difference was statistically significant for the MOMP peptide (P,0.05), but not for the Omp5 peptide when compared with the combined Cpn peptide (Figures 6C and D).

Level of Atheroprotective and Atherogenic Cytokines in Plasma and in Supernatants of Stimulated Splenocytes
Plasma levels of atheroprotective cytokine IL-10 (P,0.05) and TGF-b (P,0.01) were significantly increased in mice immunized with the combined Cpn peptide compared with either the MOMP peptide or the Omp5 peptide ( Figures 7A and B). The levels of these two cytokines were also significantly increased in mice immunized with either the MOMP peptide (P,0.05) or the Omp5 peptide (P,0.05) (Figures 7A and B) versus controls. Plasma levels of TNF-a were significant reduced by immunization with the combined peptide compared with either the MOMP (P,0.05) or the Omp5 peptide ( Figure 7C); all peptides produced a reduction in TNF-a secretion versus controls. A similar trend was obtained for these peptides in respect of plasma levels of IFN-c ( Figure 7D). Although no statistically significant difference was found between the combined Cpn peptide and the MOMP peptide, a significant difference in IFN-c levels was found in mice immunized with the combined Cpn peptide compared with the Omp5 peptide (P,0.05) ( Figure 7D).
In addition, supernatants of the splenocytes from mice immunized with these peptides individually showed significantly increased secretion of IL-10 ( Figure 7E) and TGF-b ( Figure 7F), stimulated with 10 or 100 mg/mL of Con A (P,0.001). The incremental secretion of IL-10 and TGF-b in the combined Cpn peptide-immunized mice was significantly higher than either the MOMP or the Omp5 peptide-immunized mice (P,0.05). Significantly decreased secretion of TNF-a ( Figure 7G) and IFNc ( Figure 7H) was found in the supernatants of splenocytes in mice immunized with these peptide antigens. A significantly greater decrease in the secretion of TNF-a was observed in mice immunized with the combined Cpn peptide versus immunized with either the MOMP peptide or the Omp5 peptide alone ( Figure 7G; P,0.05) stimulated with 100 mg/mL of ConA (P,0.05). A significantly greater decrease in secretion of IFN-c was detected in the combined Cpn peptide-immunized mice versus those in either the MOMP-peptide or the Omp5 peptideimmunized mice ( Figure 7H, P,0.05), stimulated with 10 or 100 mg/mL of ConA (P,0.05).
Furthermore, IL-10 concentration was significantly higher in the supernatant of spleen cells from the combined Cpn peptideimmunized mice stimulated by the same antigen peptide than that after stimulation by PBS (P,0.05, Figure 7I). In contrast, IFN-c was significantly lower in supernatant of spleen cells from the combined Cpn peptide-immunized mice stimulated by the same antigen peptide than that of mice stimulated by PBS (P,0.05, Figure 7J). In the similar experimental conditions, supernatant of spleen cells from the naive mice failed to show these differences.

Evaluation of Antigen-Induced Specific Treg Function and Specific Cellular Immune Response
To assess whether functional Treg cells were induced by immunization, we co-cultured antigen-specific Treg cells with CD4 + effector T-cells (CD4 + CD25 -T-cells). Proliferation of effector T-cells from control mice immunized with KLH in response to stimulation with KLH at 1 mM did not show suppression in a dose-dependent manner in the presence of Treg cells from KLH-immunized mice (Figures. 8A, B and F). In contrast, Inhibition of effector T-cell proliferation was achieved from sampling mice immunized with either the MOMP peptide ( Figures 8A, C and F  as well as the combined peptide ( Figures 8A, E and F), when cocultured CD4 + CD25 2 effector T cells with CD4 + CD25 + Treg cells isolated from these mice in response to stimulation with related antigen. The differences were significant when adding Tregs to the effector cells at the ratios between 4:1,16:1 (P,0.05,,0.001) compared with that without the addition of Tregs. Additionally, the differences were significant when compared different concentration of added Treg in sampling mice at the ration between 4:1,16:1 of Treg and T effector cells (P,0.05,,0.001).

Discussion
In this study, we investigated the effect of the combined peptide derived from Cpn proteins of MOMP and Omp5 in modulating experimental atherosclerosis in Apob tm2Sgy Ldlr tm1Her /J mouse-a strain that develops high levels of atherosclerosis after being fed a high-fat diet [16]. The effect of immunization with the combined peptide was compared with that of the two peptides used singly. The two peptides were selected for coupling based on our hypothesis that they share high homology with ApoB protein which is associated with atherosclerosis.
In line with previous reports by Blessing et al [17] and Ezzahiri et al [19], our results showed that infection of mice with Cpnbacteria promotes the formation of atherosclerotic lesion. Unlike that reported by Blessing et al, our results did not show statistical significance on promoting the lesion in mice infected with Cpn bacteria compared with that of uninfected mice following feeding with a high-fat diet for 10 weeks. These results suggest that feeding with a high-fat diet for longer than 10 weeks may be necessary to form sufficient lesion area, as the data from Blessing et al have shown that the fast lesion increasing time seems between 10-16 weeks [17]. Additionally, other factors, such as different mouse or bacterial strains and different inoculating doses may also affect lesion formation. We report the novel finding that either the MOMP peptide or the Omp5 peptide significantly reduces atherosclerotic lesion when used as a peptide antigen in Apob tm2Sgy Ldlr tm1Her /J mice infected with Cpn bacteria. A different study was shown that a recombinant protein containing both of the MOMP peptide and the Omp5 peptide sequence significantly reduced atherosclerotic lesion in non-infected mice [11], indicating infection of mice with Cpn is an independent factor for functional study of these two peptides. The mechanism of this function may be due to molecular mimicry, as the combined Cpn peptides can cross-react with antiserum induced by human ApoB peptide (also called ApoB100 661-681 peptide). Although certain level of cross-reaction was observed between ApoB peptide and Cpn peptide antisera, strong cross-reaction (approx net 1.5 OD value) between Cpn peptide and ApoB peptide antisera was striking. In fact, autoantibodies to this ApoB peptide as well as to other ApoB100 peptides are present in human plasma and are associated with decreased cardiovascular risk. In addition, immunization with this peptide reduces atherosclerotic lesion in several knock-out mouse models including LDL receptor 2/2 / human ApoB-100 mice [20], ApoE 2/2 mice [21] and Apob tm2Sgy Ldlr tm1Her /J mice [16]. Vaccination with a modified Streptococcus pneumoniae (Spn) has been reported to decrease atherosclerotic lesion formation through molecular mimicry between Spn and oxidized low-density lipoprotein (oxLDL) [22], which has been demonstrated to play a role in the development of atherosclerotic lesions [23,24]. Immunization of oxLDL or apolipoprotein B (ApoB) peptide of oxLDL reduces atherosclerotic lesion formation [16,25] in either ApoE 2/2 or Apob tm2Sgy Ldlr tm1Her / J mouse models. However, molecular mimicry can occur in the absence of any true sequence homology between Cpn-derived peptide and ApoB peptide by using the computer-based software 'SIM-Alignment Tool' for protein or peptide sequences (http://web.expasy.org/tools/sim), in agreement with the report by Kohm et al [26]. In this case, it could explain why antigenic surfaces, rather than sequence homology, dictate molecular mimicry [27]. The data from our present study also demonstrate that reduction of the lesion, either in the aortic sinus or the descending aorta, in mice immunized with the combined Cpn peptide is significantly greater than with either the MOMP-or the Omp5peptide alone. In addition, the reduction in lipid level at lesion sitesin mice immunized with the combined Cpn peptide is significantly greater than that in mice immunized with either the MOMP or the Omp5 peptide alone. Furthermore, the reduction of atherosclerotic lesion in mice treated with these peptides was associated with an increase in collagen content. Collagen content in mice immunized with the combined Cpn peptide is significantly greater than that in mice immunized with either the MOMP-or the Omp5-peptide alone. These data suggest that the Cpn-derived combined peptide may have an additive effect.
Immunization Keyhole limpet hemocyanin (KLH) with the combined Cpn peptide was associated with intracellular responses that influence cellular infiltration into atherosclerotic lesions, as the levels of macrophages, activated CD4 + T cells and dendritic cells (markers of early lesion formation) were decreased compared with those in controls [28]. In addition, CD4 T-cells from the mice immunized with the combined Cpn peptide harbor 2-fold more Foxp3 + Treg cells compared with either the MOMP peptide or the Omp5 peptide and approximately 20-fold more compared with controls. As Th2-type IgG1, and not Th1-type IgG2a, anti-Cpn peptide antibody titers were observed, indicating an association of Cpn peptide treatment with Th2 response which was enhanced by Treg [29,30].
It appears that there is a shift toward decreased proinflammatory cytokine secretion and increased anti-inflammatory cytokine production in antigen-immunized mice, as the main biological function of IL-10 and TGF-b is to limit and terminate the inflammatory responses [31]. IL-10 has been reported to be a global suppressor of immune responses as well as an immunoregulator of the Th-2 cell response [31] while IL-17A (previously called IL-17) is a unique T helper lineage that regulates tissue inflammation [32]. Evidence was obtained from CD4 + expressing pro-atherogenic cytokine IL-17A [33,34], which was shown decreased IL-17A population in peptide antigen-immunized mice in this study.
The suppressive function of CD4 + CD25 + Foxp3 + natural Tregs appears to require only cell-cell contact or proximity in vitro, whereas the in vivo function of these cells is associated with secretion of IL-10 and TGF-b [11,35]. We demonstrated in vitro that the atheroprotective effect paralleled an induction of Treg suppression of antigen-specific effector T-cells, thus suggesting Tregs have an active role in the control of the atherosclerotic process. Treg cells are characterized by the expression of Foxp3, which has a crucial role in their suppression function [36]. Collectively, our study has provided evidence that the mechanism of lesion reduction by subcutaneous immunization with Cpn-derived hepta-or nona-peptide is duo to antigen-induced specific Treg expansion, which suppresses T effector cell proliferation along with increased atheroprotective cytokines and decreased proinflammatory cytokines. Notably, IL-17A that was thought to play a pro-atherogenic role in atherosclerosis [37,38] was down-regulated, however, the relevance of IL-17A to human atherosclerosis remains poorly defined.
In conclusion, we have provided proof of principle evidence that the combined Cpn peptide antigen has an additive effect as this peptide in most of experiments, if not all, showed statistically significant difference when compared to either the MOMP-or the Omp5-peptide antigen used singly (Table S3). It appears that modulation of atherosclerosis-related autoimmunity by antigen-specific activation of Tregs represents a novel approach for the development of bivalent vaccines against atherosclerosis.  Figure S3 Assessment of IL-17A expression level in splenocytes from Ldlr tm1Her Apob tm2Sgy J mice infected with Cpn bacteria and fed a high-fat diet after immunization with Cpn peptides vs controls infected with Cpn bacteria and fed a high-fat diet after immunization with KLH only. A. Representative flow cytometry plots for IL-17A expressing CD4+ population in spleen cells. Spleen cells from mice infected with Cpn bacteria and fed a high-fat diet after immunization with Cpn peptides and control mice infected with Cpn bacteria and fed a high-fat diet after immunization with KLH only were purified using a CD4 + purification kit (Miltenyi Biotec, Surrey, UK) according to manufacturer's protocols. B. Bar chart presentation of flow cytometry analysis. Data represent mean 6 SEM from 3 independent samples. (TIF)