Mycophenolate mofetil decreases atherosclerotic lesion size by depression of aortic T lymphocyte and IL-17-mediated macrophage accumulation

Abstract

Objective

This study tested whether immunosuppression with mycophenolate mofetil (MMF) inhibits atherosclerosis development in apolipoprotein-E–deficient (Apoe^−/−) mice and investigated the mechanism.

Background

Chronic vascular inflammation involving both innate and adaptive immunity is central in the development of atherosclerosis, but immunosuppressive treatment is not uniformly beneficial. The immunosuppressive MMF targets lymphocyte proliferation by inhibiting inosine-monophosphate dehydrogenase.

Methods

Young and aged Apoe^−/− mice were treated with 30 mg/kg*day MMF during 12 and 3 weeks of high fat diet, respectively. Aortic lesion size and composition was investigated by histology and flow cytometry, soluble inflammatory mediators were investigated by ELISA.

Results

Macroscopic and histologic aortic atherosclerotic lesions were significantly decreased in both MMF-treated groups. While systemic IgG directed against low density lipoproteins was not significantly altered, the T cell cytokine interleukin (IL)-17 was significantly reduced in plasma of MMF-treated mice and supernatants from their aortas after T cell stimulation. MMF treatment decreased aortic αβTCR⁺ lymphocyte proliferation and cell numbers. Also, aortic content of CD11b⁺CD11c⁺ cells and their proliferation were reduced in MMF-treated Apoe^−/− mice. IL-17 supplementation restored the number of proliferating aortic CD11b⁺CD11c⁺ cells in MMF treated mice. IL-17 receptor A was highly expressed on circulating monocytes that are macrophage progenitors. Genetic deletion of IL-17 receptor A or IL-17A reduced inflammatory peritoneal CD11b⁺CD11c⁺ macrophage accumulation.

Conclusions

The lymphocyte-directed immunosuppressant MMF that curbs IL-17 production was a successful anti-atherosclerotic treatment. Our data delineate a role for IL-17 in CD11b⁺CD11c⁺ cell accumulation.

Introduction

Inflammatory leukocyte immigration into the vessel wall, proliferation and differentiation is central during atherosclerotic plaque formation (1), the major cause of death worldwide (2). While foam cell forming macrophages are the most prominent cells, lymphocytes are present in the atherosclerotic intima, media and adventitia in humans (3) and mice (4) and proliferate in human atherosclerotic arteries (5). Dissection of adaptive immune components in genetic deletion models provides evidence that B cells protect against atherosclerosis development by production of antibodies directed against oxidized LDL (6–8). The effect of T cells on atherosclerosis formation depends on the T cell subtype. T_H1 cells and the T_H1 marker interferon (IFN)-γ promoted atherosclerotic disease in a large number of models, the role of T_H2 cells is not clear (9,10) and anti-inflammatory regulatory T cells and their key signaling molecules limited atherosclerotic lesion formation (11,12). The role of T_H17 cells that produce interleukin (IL)-17, (also called IL-17A(13)) in atherosclerosis is currently under investigation. Transplantation of LDL-receptor deficient (Ldlr^−/−) mice with IL-17-receptor-A deficient (Il17ra^−/−) bone marrow decreased lesion size (14) as did blockade of IL-17 receptor signaling (15) or IL-17A in Apolipoprotein E deficient (Apoe^−/−) mice (16) but IL-17 injection into Ldlr^−/− mice decreased atherosclerotic lesion size according to another report (17).

Pharmacologic intervention during atherosclerotic lesion formation with systemic immunosuppression yielded diverse and rarely beneficial effects. Cyclosporin A at a dose that affected T cell, but not B cell function enhanced atherosclerotic lesion development in cholesterol-fed rabbits (18). FTY720 reduced atherosclerosis in Apoe^−/− (19) and Ldlr^−/− (20) mice on high fat diet, but also increased circulating lipid levels in Apoe^−/− mice on chow diet (21). Mycophenolic acid (MPA) is a purine antagonist that acts by inhibition of inosinemonophosphate dehydrogenase (IMPDH) and thereby blocks de novo generation of guanosine nucleotides required by proliferating cells, while other cells use a salvage pathway (22). MPA preferentially targets the IMPDHII isoform that is mainly expressed in activated B and T lymphocytes (22,23). MPA effects on vascular disease in transplanted organ allografts were favorable compared to calcineurin inhibitors, but also compared to 6-mercaptopurine, another purine antagonist (24–26). Beneficial effects of MPA on native vessel atherosclerosis have been proposed (27,28) but experimental data is limited to two short reports describing decreased intimal thickness in rabbits that were given either 30mg/kg of the prodrug mycophenolate mofetil per gavage (29) or 80mg/kg subcutaneously (30) during 12 weeks of high fat diet. No mechanism was investigated.

This study tested the impact of MMF, an immunosuppressant approved for use in humans that is comparatively well tolerated, on atherosclerotic lesion formation in Apoe^−/− mice. We characterized the aortic leukocyte infiltrate and cytokine production to determine mechanisms of MMF’s atheroprotective effects.

Methods

Animals

Wild-type (wt) C57Bl/6 mice, Apoe^−/− mice (Jackson Labs, Bar Harbor, ME), mice lacking IL-17A (Il17a^−/−), (Dr. Iwakura, Tokio), or IL-17-receptor-A (Il17ra^−/−), (Dr. Peschon, Amgen), (both 96% C57Bl/6 background, males and females) were genotyped by PCR and used in age and sex-matched groups. Animal numbers for each specific analysis are given in the figure legends. Animal experiments were approved by the Animal Care Committee at LIAI.

Mycophenolate as the prodrug mycophenolate mofetil (Roche Pharma AG, Grenzach-Wyhlen, Germany) was incorporated into high fat diet without cholate (405 of kcal from fat, 1.5% Cholesterol) (Research Diets, New Brunswick, NJ) to reach an oral dose of 30mg/kg body weight*day at a chow consumption of 0.1g/g body-weight. Monitored chow consumption was 2.9±0.09g (n=33 daily measurements at a body weight of 21.7±1g (n=8 mice), resulting in an average consumption of 0.13 g/g*day. Addition of 10% plasma from treated mice inhibited T cell proliferation to 59.9±8% of control Apoe^−/− mice on high fat diet plasma (n=6), representing a plasma concentration of 1.6±1.4 µg/ml compared to a standard curve as described (31).

Blood for leukocyte counts analyzed by automatic analyzer (Hemavet 950FS, DREW Scientific, Oxford, CT). Total plasma cholesterol was analyzed using resorufin fluorescence according to the manufacturer’s instructions (Cayman, Ann Arbor, MI). Recombinant IL-17A was from Peprotech (Rocky Hill, NJ).

Quantification of atherosclerosis and histologic analysis

Aortas were excised, fixed and stained with SudanIV (counterstain fast green/hematoxylin)(32). Digital images were obtained using moticam 1000 (Motic, Richmond, Canada) on an Olympus S267 dissection scope (Olympus, Center Valley, PA). 5µm sections of aortic roots were starting at the aortic valve plane and covering 300µm in 50µm intervals were used for histologic lesion size quantification (Photomicrographs taken with a 4x objective/Nikon eclipse 80i microscope, SudanIV/hematoxylin/light-green stain). Lesion size was determined using NIH Image J and averaged over all sections per mouse. For immunofluorescence, purified rat-anti-CD3! (17A2)(eBioscience) was used with goat anti-rat-Alexa488 (invitrogen). Images were acquired on a Leica DM6000 upright microscope with DIC optics using a HCX PLAPO 20x oil-immersion objective at 488 nm excitation wavelength. NIH Image J was employed to adjust brightness and for one-step smoothing.

Enzymatic digestion of tissues and flow cytometry

After sacrifice and perfusion (PBS/20 U/ml heparin), complete thoracic and abdominal aortas were freed of all visible adventitial fat at 4× magnification and digested as described (33). T cell stimulation was conducted with plate-bound purified anti-CD28 and anti-CD3 (Biolegend, San Diego, CA) in full RPMI. Flow cytometry analysis was performed on a Becton-Dickinson LSRII, data analyzed using FlowJo software (Tree Star Inc., Ashland, OR). For aorta analysis, all events were acquired, 500,000 events were read for all other organs. Gating was performed for live, CD45⁺ events constituting the total leukocyte population (leukocyte viability 35–60%). Antibody clone numbers are available from the authors. Aqua LIVE/DEAD® Fixable Dead Cell Stain Kit (Invitrogen, Carlsbad, CA), BD-Fix-Perm and APC BrdU flow kit (BD Pharmingen, San Jose, CA) were used according to the manufacturer’s instructions. The gate for BrdU⁺ cells was set by cells from non-BrdU− injected animals after identical preparation and antibody treatment.

Thioglycollate induced peritonitis

1ml BBL fluid thioglycollate medium (Becton-Dickinson, Sparks, MD) was injected intraperitoneally and cells were recovered after three days by washing twice with 5ml PBS.

ELISA

DuoSet Elisa development kit for IL-17A (R&D systems, Minneapolis, MN) and BD cytometric beads array mouse Th1/Th2/Th17 cytokine kits (BD biosciences) were used according to the manufacturers instructions.

Quantification of anti-native LDL, anti-MDA LDL and anti-oxLDL antibody titers

Antibody titers were determined by chemiluminescent ELISA as described (34).

Statistical Analysis

Two-tailed student t-test or Wilcoxon test was used to compare treated with untreated conditions for normally and non-normally distributed parameters (if indicated by d’Agostino Pearson test), respectively. 1-way ANOVA was used if more than two conditions were compared. P-values <0.05 were considered significant. Data are expressed as mean ±SEM. P values are indicated as *p<0.05, **p<0.01.

Results

MMF treatment is well tolerated and decreases aortic lesion size in Apoe^−/− mice

Apoe^−/− mice were maintained on high fat diet for 12 weeks. MMF at a typical clinical dose of 30mg/kg*day was well tolerated, no diarrhea, a typical side effect, occurred. Despite the fact that treated mice gained slightly less weight than controls, plasma cholesterol levels were not significantly different (table 1). MMF treatment decreased total lymphocyte counts in spleen and peripheral blood (table 1).

Table 1.

Characteristics of Apoe^−/− mice after 12 weeks of high fat diet

	ctrl	MMF	p-value
Body weight (g)	27.6±1.8 (14)	24.8±1.7 (14)	0.003**
Spleen weight (g)	0.13±0.01 (14)	0.11±0.01 (14)	0.47
Plasma cholesterol (mg/dl)	543±52 (6)	603±61 (6)	0.16
Lymphocytes/spleen	145±9.3*106 (12)	121±6.5*106 (12)	0.05*
Peripheral blood lymphocytes (103/µl)	4.77±0.48 (10)	3.44±0.28 (10)	0.03*
Peripheral blood monocytes (103/µl)	0.45±0.06 (10)	0.32±0.03 (10)	0.06

Values given as ± SEM (n), p was calculated using student’s t-test.

Atherosclerotic lesion size was determined by Sudan IV stain of whole aortas and aortic roots after 12 weeks on high fat diet. MMF significantly decreased en face atherosclerotic lesion size (figure 1A,B). Histology of the aortic roots also demonstrated a significant decrease in lesion area (figure 1C,D). While the total lesion size was decreased, the proportions of lipid or collagen staining (Sudan IV and Picrosirius red, data not shown) were not significantly altered.

Figure 1. MMF reduces aortic atherosclerotic lesion size.

Seven-week-old Apoe^−/− mice were treated with high fat diet for 12 weeks. Aortic lesion size was assessed by SudanIV staining (A,B; % of aortic surface area, n=5 (ctrl) and 7 (MMF) from two independent experiments). Aortic root lesions (300µm following the aortic valve in 50µm intervals) were significantly reduced by MMF (C, n=4). Examples of aortic root lesions are shown in D (SudanIV/light green)(*p<0.05 by student’s t-test). Abbreviations as indicated in the abbreviations table.

MMF decreases inflammatory cytokine concentration in plasma of atherosclerotic mice

To test the impact of MMF treatment on mediators of systemic adaptive immunity, plasma antibody and cytokine concentrations were studied after 12 weeks of high fat diet. Total circulating immunoglobulin IgM and IgG2c levels were unchanged, but IgG1 that is considered T_H2 associated was mildly but significantly lower (figure 2A). IgG1 and IgG2c directed against native LDL, oxidized LDL and MDA-LDL were not significantly decreased by MMF-treatment (figure 2B–D). The pro-inflammatory cytokine TNF-γ and IL-6 plasma concentrations were significantly lower in MMF-treated mice (figure 2E). Among T cell cytokines in plasma, the T_H2 marker IL-4 was below detection limit in both groups (data not shown). IFN-γ as a T_H1 marker was not significantly reduced. IL-17A as a marker of T_H17 lineage was significantly reduced (figure 2E).

Figure 2. Plasma immunoglobulin and cytokine levels in control and MMF-treated Apoe^−/− mice.

Plasma immunoglobulin levels after twelve weeks high fat diet (starting age: 7 weeks) were assessed by ELISA. Total IgM and IgG2c were unchanged, IgG1 slightly decreased in MMF treated mice (A). No significant changes were observed in IgG1 or IgG2c directed against either LDL (B–D). Plasma cytokine levels were determined by cytometric beads assay (E) (n=10 per group from 3 independent experiments, (*p<0.05 by student’s t-test)). Abbreviations as indicated in the abbreviations table.

Aortic T cell IL-17 production and leukocyte infiltration in atherosclerosis is decreased by MMF treatment

To address cytokine changes in the atherosclerotic vessel, we examined culture supernatants of aortic explants. Single cell suspensions from whole aortas were subjected to 48h T cell stimulation with anti-CD3 and anti-CD28. Subsequently, T helper subtype marker cytokine secretion into the supernatant was analyzed. Figure 3A shows that aortic IL-17, but not IFN-γ or IL-4, production was significantly decreased in aortas harvested from MMF-treated animals.

Figure 3. MMF treatment reduces aortic IL-17 production and T cell accumulation in the aorta.

T cells from aortas of Apoe^−/− mice after three weeks high fat diet with and without MMF (starting age: 7 weeks) were stimulated with anti-CD3 and anti-CD28 for 48h. Cytokines in the supernatants were analyzed by cytometric bead assay (A, n=5 (ctrl) and 6 (MMF)). Aortic leukocytes were assessed by flow cytometry (B). Total cell counts and αβTCR⁺ cells were significantly reduced by MMF treatment (C, n=8 (ctrl) and 10 (MMF) from 3 independent experiments) (*p<0.05 by student’s t-test). Abbreviations as indicated in the abbreviations table.

We examined the effects of oral MMF treatment on aortic leukocytes in Apoe^−/− mice at three weeks on high fat diet. At this time point, atherosclerotic lesions became visible as fatty streaks, mainly in the in-nominate artery. Splenic and circulating lymphocyte counts were very similar in MMF-treated and control mice (data not shown). Leukocyte quantification was performed by flow cytometry. Figure 3B,C shows a significant decrease in all leukocytes and αβTCR⁺ cells in the aortas of MMF-treated mice. We also tested whether MMF treatment was effective in aged mice with preexisting disease. 20–22 week old Apoe^−/− mice were given high fat diet with or without MMF for three weeks. Again, T cell infiltration into the aorta quantified by flow cytometry was significantly smaller in the MMF treatment group (figure 4A). Both macroscopic aortic arch (Figure 4B) and histologic lesions (figure 4C,D) were smaller in MMF-treated mice.

Figure 4. MMF reduces atherosclerotic lesion size and aortic T cell infiltration in aged Apoe^−/− mice.

20–22 week old Apoe^−/− mice were given three weeks high fat diet with and without MMF. Flow cytometry analysis was performed for quantification of aortic αβTCR⁺ cells (A). Macroscopic aortic arch lesion size was smaller in MMF-treated than control mice (B, typical examples). Histologic assessment of aortic roots revealed a significantly decreased lesion size (C, typical examples and D, statistical analysis) (n=4 (ctrl) and 5 (MMF) from 2 independent experiments, *p<0.05 by student’s t-test). Abbreviations as indicated in the abbreviations table.

MMF treatment decreases the number of proliferating T lymphocytes in the aorta during development of atherosclerosis

MMF inhibits lymphocyte proliferation. Aortic leukocyte proliferation is highest early in atherosclerosis development (35). Cell proliferation was assessed 24h after intraperitoneal BrdU injection after three weeks of high fat diet with and without MMF (figure 5). Proliferation in αβTCR⁺ cells recovered from the aorta was markedly higher than in spleen where less than 1% of αβTCR⁺ cells proliferated within 24h (figure 5A,B). The low level of proliferation in spleen was not significantly altered by MMF at 30mg/kg as employed in our trial. The amount of αβTCR⁺ cells in the aorta that proliferated was significantly decreased by MMF (figure 5C,D).

Figure 5. MMF treatment reduces aortic lymphocyte proliferation during the development of atherosclerosis.

After three weeks high fat diet (starting age: 7 weeks) with and without MMF, Apoe^−/− mice were injected with BrdU 24h prior to sacrifice. BrdU incorporation was detected by subsequent intracellular staining of single cell suspensions from spleen (A,B) and aorta (C,D). Gates were set by identically-prepared cells from the same organ of mice not given BrdU. Aortic T cell proliferation was significantly decreased by MMF treatment (D, n=8 (ctrl) and 10 (MMF) from 3 independent experiments, *p<0.05 by student’s t-test). Abbreviations as indicated in the abbreviations table.

Less aortic T cells in MMF treated mice with established atherosclerosis after 12 weeks high fat diet

Leukocytes from aortas of mice with established atherosclerosis after 12 weeks high fat diet were analyzed. Similar to the early time point, both the percentage among all leukocytes and the absolute number of infiltrating αβTCR⁺ cells were significantly decreased by MMF (figure 6A,B). The proportions of CD4⁺CD25⁺ cells, which include regulatory T cells, in aorta and spleen were unaltered (data not shown) arguing against a major impact of MMF on this cell type in this model. Immunofluorescence was used for localization of T cells within the aortic wall. CD3⁺ T cells were found in clusters in the adventitia but also in the plaque. In aortas of MMF-treated mice, CD3⁺ T cell numbers were decreased and adventitial clusters dispersed (figure 6C).

Figure 6. Decrease in aortic T cells is sustained in Apoe^−/− mice after 12 weeks on high fat diet with MMF.

Flow cytometry was used for quantification of leukocyte infiltration after 12 weeks high fat diet (starting age: 7 weeks). The proportion of αβTCR⁺ T cells among all leukocytes (A, typical example) and the absolute number of αβTCR⁺ T cells per aorta (B) were significantly reduced by MMF (n=8 from 3 independent experiments, *p<0.05 by Wilcoxon test). Immunofluorescence of frozen aortic root sections after 12 weeks on high fat diet revealed decreased CD3-staining in the lesions of MMF-treated as compared to control mice (C, 20× original magnification, bars represent 100µm, isotype and secondary antibody served as specificity controls). Abbreviations as indicated in the abbreviations table.

MMF treatment alters aortic macrophages indirectly via depression of IL-17

We further investigated aortic macrophages, that are central in atherosclerosis development. While MMF did not significantly decrease the total number of CD11b⁺ macrophages, approximately half of aortic CD11b⁺CD45⁺ leukocytes also expressed the β2 integrin subunit CD11c after 12 weeks of high fat diet in control Apoe^−/− mice. In MMF-treated animals, there were less CD11b⁺CD11c⁺ than CD11b⁺CD11c⁻ cells among all CD11b⁺ cells (figure 7A). The mean fluorescence intensity reflecting the number of CD11b and CD11c molecules on the surface of each individual cell in the aortic CD11b⁺CD11c⁺ population remained unaffected by MMF treatment (data not shown), suggesting that the observed effect was a decrease in a cell population rather than relative integrin surface expression. CD11c is expressed on CD11b⁺ cells in atherosclerotic plaques and up-regulated by oxidized LDL (35,36). CD11c-deficiency decreased atherosclerotic lesion size (37). We assessed proliferation of aortic CD11b⁺CD11c⁺ cells after 3 weeks on high fat diet. It was significantly decreased by treatment with MMF (figure 7B).

Figure 7. MMF-mediated decrease of aortic CD11b⁺CD11c⁺ macrophage proliferation is rescued by IL-17 and IL-17 promotes CD11b⁺CD11c⁺ cell accumulation in vivo.

After twelve weeks on high fat diet, CD11c⁺ among CD11b⁺ leukocytes were reduced in aortas of MMF-treated compared to control Apoe^−/− mice (A). Aortic CD11b⁺CD11c⁺ cell proliferation measured by BrdU incorporation was reduced after three weeks high fat diet (B, n=8 (ctrl) and 10 (MMF) from 3 indep. experiments, starting age 7 weeks, *p<0.05 by student’s t-test,).

To test whether the decrease in CD11b⁺CD11c⁺ cell proliferation was due to inhibition of IL-17 production by MMF, recombinant IL-17 (1µg/mouse) or PBS control was injected intraperitoneally on days 5, 10 and 15 of high fat diet with MMF. Aortic CD11b⁺CD11c⁺ cell proliferation is depicted in C (n=9 (ctrl) and 10 (MMF) from two independent experiments, starting age 7 weeks, *p<0.05 by student’s t-test). In-nominate artery lesions at three weeks of western diet were reduced by MMF but restored in IL-17-substituted MMF-treated Apoe^−/− mice (D, typical examples).

Interleukin-17-receptor expression on peripheral blood CD11b⁺CD115⁺ monocytes was analyzed by flow cytometry, monocytes from Il17ra^−/− mice served as control (E). Peritoneal macrophages from wild-type (wt), Il17a^−/− and Il17ra^−/− mice were harvested on day 3 after thioglycollate injection. The number of CD11b⁺CD11c⁺ macrophages per peritoneal cavity was lower in Il17a^−/− and Il17ra^−/− than wild-type mice (F, n=4 from independent experiments, Dunnett’s test after 1-way ANOVA, *p<0.05). Abbreviations as indicated in the abbreviations table.

To test whether the decreased proportion of CD11b⁺CD11c⁺ macrophages in MMF-treated mice aortas was a direct or a downstream effect due to decreased IL-17 production, Apoe^−/− mice were injected with either recombinant IL-17 or PBS control on day 5,10 and 15 of a three week course of high fat diet and MMF. This completely restored CD11b⁺CD11c⁺ proliferation in the aorta (figure 7C). At this point, atherosclerotic lesions were minimal in the aortic root (data not shown), but macroscopically visible the branching in-nominate artery. These were reduced in MMF treated mice, and this was reverted by IL-17 treatment (figure 7D). Indeed, IL-17-receptor-A was highly expressed on circulating CD11b⁺CD115⁺ monocytes (figure 7E) that are precursors of aortic macrophages (35,36). To address whether IL-17 had a role in inflammatory macrophage accumulation in circumstances other than MMF treatment, we employed IL-17A (Il17a^−/−) deficient and Il17ra^−/− mice. Baseline circulating monocyte counts were not significantly different from wildtype mice (data not shown). Thioglycollate-induced peritoneal macrophages were harvested after 3 days. Less CD11b⁺CD11c⁺ cells were recovered from the peritoneal cavity of both Il17ra^−/− and Il17a^−/− mice than from wild-type controls (figure 7F).

Discussion

Targeting lymphocyte proliferation by inhibition of inosine-monophosphate dehydrogenase by MMF reduced atherosclerotic lesion formation in Apoe^−/− mice. This was achieved both during 12 weeks of high fat diet in young or during 3 weeks in aged mice.

MMF effects on B and T cells and T cell cytokine production during development of atherosclerosis

MMF targets proliferating lymphocytes (22). We observed no significant change in IgG, which are considered to mediate atheroprotection by B cells (6–8). Concerning T lymphocytes, MMF treatment predominantly reduced aortic αβTCR⁺ cell accumulation. T cell numbers were reduced in both plaque and adventitia. The proportion of proliferating T cells was much higher in aortas with nascent atherosclerotic lesions than in the spleen of the same animal. Aortic T cell proliferation was preferentially affected by MMF. Subsequently, both after 3 and 12 weeks high fat diet, MMF treatment had a stronger effect on aortic lymphocyte populations than in spleen or peripheral blood. A very recent report suggests that similar effects can be observed in humans: T lymphocytes in carotid endarterectomy specimens from patients treated with MMF for 2 weeks prior to surgery were decreased (38).

Certain T cell populations, especially T_H1 and possibly T_H17 cells, have a proatherogenic role (1,11,39). Inhibition of T cell proliferation during an inflammatory event such as atherosclerotic lesion formation can reduce T cell cytokines by decreasing the number of cytokine producers. Indeed, systemic and aortic IL-17 cytokine levels were lower in MMFtreated than control mice. Our present experiments do not explore why IL-17 producing T cells were preferentially affected by MMF treatment. However, from the mechanism of action of MMF it is conceivable that these cells were most affected because they or their progenitors were proliferating most actively, making them most vulnerable to guanosine nucleotide depletion similar to the setting of bone marrow regeneration as we recently explored (31).

The role of IL-17 in atherosclerosis development

Most, although not all, current literature suggests a pro-atherogenic role of IL-17 (14–17,40) by a yet undefined mechanism. Our data do not exclude that additional MMF effects beyond IL-17 suppression have contributed to its beneficial effects. However, we investigated whether IL-17 was involved in the effects of MMF-mediated T cell suppression on atherosclerosis by substituting IL-17 into MMF treated mice. This reverted the MMF effect on nascent lesion formation and on a cellular level on aortic CD11b⁺CD11c⁺ macrophages. These cells are expanded in atherosclerosis (35,36), present antigens to CD4⁺ T cells and phagocytose lipids (39,40). IL-17 receptor was highly expressed on circulating monocytes that are progenitors of aortic macrophages. We tested for a direct link between IL-17 and CD11b⁺CD11c⁺ cells in acute peritoneal inflammation in mice lacking IL-17A or IL-17-receptor-A. The number of CD11b⁺CD11c⁺ cells was reduced in both genotypes. Our data suggest IL-17 as a mediator of CD11b⁺CD11c⁺ cell accumulation. Also aortic CD11c⁺ number was lower in Il17ra^−/− mice on high fat diet compared to wildtype controls (unpublished observation), suggesting relevance in vascular inflammation. IL-17 increased monocyte adhesion to the atherosclerotic aortic wall ex vivo (15), but it remains to be determined at which step IL-17 promotes CD11b⁺CD11c⁺ cells. The MMF-induced decrease in CD11b⁺CD11c⁺ cells may have contributed to systemically lower TNF-α and IL-6 inflammatory cytokine levels (figure 2) and decreased numbers of neutrophils in early atherosclerotic aortas (data not shown). IL-6 can again induce T cell IL-17 production (41), a pro-inflammatory loop that was tempered in MMF treated mice.

Immunosuppression in atherosclerosis management

Systemic inhibition of T cell immunity in native vessel atherosclerosis with other agents was pro-atherogenic for cyclosporine (18) or only partially successful for FTY720 (19–21). Our finding that MMF conferred protection not only to young, but also aged mice is promising. Further studies are necessary to delineate whether, in addition to inhibiting new plaque formation, MMF also decreased pre-existing disease in aged mice. Systemic immunosuppression with any of the currently available agents including MMF despite a relatively favorable side effect profile (22,42) confers major risks to patients. These preclude their chronic use for atherosclerosis as the sole indication, however, beneficial vascular side effects might influence the choice of immunosuppressant prescribed for other conditions in patients with high cardiovascular risk. Also, our finding that the alteration in CD11b⁺CD11c⁺ cells in aortas of MMF-treated atherosclerotic mice depended on IL-17 suggests IL-17 as a mediator and possibly more specific target for immunosuppressive therapy of atherosclerosis.

Abbreviations

MMF

mycophenolate mofetil

MPA

mycophenolic acid

Apoe

Apolipoprotein E

IL-17

Interleukin 17A

IL-17ra

IL-17 receptor A

TCR

T cell receptor

BrdU

Bromodeoxyuridine