| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Atherosclerosis and Lipoproteins |
From the Department of Cardiology (R.K., F.B., E.B., M.P., H.A.K., T.J.D.), University Hospital Heidelberg, Germany; Department of Medicine (R.K., G.K.H.), Karolinska Institutet, Stockholm, Sweden; the Leibniz-Institute for Arteriosclerosis Research (J.R.N.), University of Münster, Germany; the Center for Endocrinology, Metabolism, and Diabetes (M.R.), Karolinska Institutet, Stockholm, Sweden; and Deutsches Krebsforschungszentrum (DKFZ) (H.J.G.), Heidelberg, Germany.
Correspondence to Roland Klingenberg, MD, Center for Molecular Medicine, L8:03, Experimental Cardiovascular Research Unit, Karolinska Hospital, S-171 76 Stockholm, Sweden. E-mail roland.klingenberg{at}ki.se
| Abstract |
|---|
|
|
|---|
Methods and Results— ApoE–/– mice on a normal laboratory diet received oral FTY720 for 12 weeks, which led to a 2.4-fold increase in serum cholesterol (largely VLDL fraction) and a 1.8-fold increase in hepatic HMGCoA reductase mRNA. FTY720 increased plasma sphingosine-1-phosphate and induced marked peripheral blood lymphopenia. A discoordinate modulation of B, T and monocyte cell numbers was found in peripheral lymphoid organs. Overall depletion of T cells was accompanied by a relative (2-fold) increase in regulatory T cell content paralleled by a similar increase in effector memory T cells (CD4+CD44hiCD62lo) as absolute numbers of both subpopulations remained essentially unchanged. Lymphocyte function was unaltered as indicated by anti-OxLDL antibodies and T cell proliferation. There were no changes in atherosclerotic lesions in early and established atherosclerosis.
Conclusions— FTY720 mediated peripheral lymphocyte depletion and retrafficking without altering function and overall balance of pro- and antiatherogenic lymphocyte populations. A net decrease in lymphocyte numbers occurred concomitantly with a more proatherogenic hypercholesterolemia resulting in unaltered atherogenesis.
Administration of the sphingolipid analogue FTY720 to ApoE–/– mice on normal laboratory diet altered lipid metabolism yielding pronounced hypercholesterolemia. Despite lymphocyte depletion and retrafficking the overall balance of pro- and antiatherogenic lymphocyte populations was not changed. Hypercholesterolemia appears to have counteracted the otherwise beneficial effect on atherogenesis.
Key Words: atherosclerosis immune system immunosuppressive therapy leukocytes vascular biology
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
Animals
Male ApoE–/– mice were obtained from Jackson Laboratories (Bar Harbor, Me). Ethical permission was obtained from the University of Heidelberg. Mice were fed a normal laboratory diet (ssniff Spezialdiäten GmbH). FTY720 (Novartis) was continuously administered for 12 weeks to the animals by drinking water calculated at a daily dose of 3 mg/kg/d.7 8-week-old mice (n=7/group) and 26-week-old mice (n=10/group) were treated for 12 weeks with FTY720 to evaluate lesions in the aortic root20 and innominate artery.21
Tissue Processing
For RNA isolation the aortic arch, a predilection site for lesion development in ApoE –/– mice,22 was dissected and snap-frozen. Lesion area and fractional area of the lesion were quantified and the results were expressed as the average of 8 sections per mouse.20
Immunohistochemistry
Light microscopy was performed on 10-µm cryosections adjacent to the ORO-stained sections. Primary antibodies (CD4, CD8, CD19, CD68, vascular cell adhesion molecule [VCAM]-1, I-Ab; all rat anti-mouse by BD Pharmingen) were titrated to optimum performance and applied to acetone-fixed cryosections followed by detection with the ABC alkaline phosphatase kit (Vector Laboratories). A thresholding technique using computerized ImagePro analysis on the aortic root sections was implemented.
Flow Cytometry
Flow cytometry was performed on a CyAn (Dako) after staining with the appropriate Ab; data were analyzed using Summit v4.3 software. Primary labeled antibodies used were CD19, CD3, CD4, CD8, F4/80, CD25, CD62L, CD44 from Pharmingen, FoxP3-PE was from eBioscience.
Functional Immune Assays
Splenocytes were harvested and cultured in duplicate in a 96-well plate at 5x105 cells per well after red blood cells lysis. Cells were incubated for 72 hours in the absence or presence of anti-CD3 antibody (1 µg/mL) followed by incorporation of 3H-thymidine during the last 18 hours. ELISA methods were used to quantitate serum Ig isotypes to Ox-LDL.23
Real-Time Polymerase Chain Reaction
RNA was isolated from the aortic arch using the RNeasy kit (Qiagen). Reverse transcription was performed using the Boehringer cDNA kit (Roche Diagnostics). The Roche real-time polymerase chain reaction (PCR) kit with SYBR Green (Roche Diagnostics) was used for quantitative PCR (LightCycler). Primer sequences were previously published.24–28 Data were analyzed on the basis of the relative expression method with the formula 2–
CT, where 
CT=
CT (sample)–
CT (calibrator=average CT values of all samples), and
CT is the CT of the housekeeping gene (β-actin) subtracted from the CT of the target gene.
Hematologic and Biochemical Parameters
Whole blood (EDTA) was analyzed by Cell Dyn 3500 hemocounter (Abbott). Serum total cholesterol and triglycerides were determined using a Monarch Automated Analyzer (ILS Laboratories Scandinavia AB). Fast protein liquid (FPLC) (Amersham Pharmacia) was performed for detection of cholesterol.29 FTY720 serum levels were determined at Novartis by high-performance liquid chromatography (HPLC). S1P plasma levels were determined as described.6 BD Cytometric Bead Array technique (Becton Dickinson and Company) was used to measure cytokine levels in serum.
Statistical Analysis
Values are expressed as mean±SEM unless otherwise indicated. Nonparametric Mann–Whitney U test was used to compare individual groups of animals. A level of P<0.05 was considered significant.
| Results |
|---|
|
|
|---|
|
Cellular Composition in Immune Organs
Two distinct lymph node sites (axillary and inguinal), spleen, and peripheral blood were analyzed for cellular composition. FTY720 mediated a redistribution of B cells (CD19+) from peripheral blood into spleen and lymph nodes whereas T cells (CD4+ and CD8+) were depleted in all 3 lymphoid tissues (Figure 2A and 2B, spleen data not shown). The percentage of monocytes was unchanged in blood and spleen, but increased in lymph nodes. Analysis of regulatory T cells (CD4+FoxP3+) and memory effector T cells (CD4+CD44hiCD62Llo) showed no difference between groups when expressed as percentage of total parenchymal cells. However, when the distribution of Treg and memory effector T cells within the CD4+ T population was examined we found a significant (2-fold) relative increase among total CD4+ T cells for both, regulatory T cells, and memory effector T cells in the FTY720 treated animals (Figure 2C through 2F). This effect was attributable to the decline in total CD4+ T cell numbers on FTY720 administration whereas absolute cell numbers for Treg and memory effector cells remained essentially unchanged (supplemental Figure II). Data on Treg and memory effector T cells in blood are not shown as total CD4+ cell content was diminished to less than 1% of total cells in the FTY720 group obviating accurate interpretation and reasonable statistical analysis because of minimal cell numbers. As both subpopulations showed the same relative increase (2-fold), the overall balance of regulatory T cells and memory effector T cells was maintained in the FTY720 treated group. These data show a discoordinate modulation of lymphocyte populations with a preserved overall balance of pro-and antiatherogenic T cells.
|
Serum Antibody and Cytokines Levels
Isotype analysis of serum antibodies against OxLDL, implicated in the pathogenesis of atherosclerosis, showed no significant differences between the 2 groups except for the subclass IgG1, which was lower in the treated group (0.28±0.03 versus 0.17±0.02, p = 0.004; Figure 3A). Serum cytokine levels for interleukin (IL)-5, IL-10, and IFN-
were similar in treated and control animals (data not shown). Anti-CD3 induced splenic T cell proliferation was not influenced by FTY720 administration as illustrated in Figure 3B. Thus, functional properties of B and T cells remained essentially unaltered by FTY720 treatment.
|
Lesion Size, Cellular Composition, and Cytokine Pattern
Lesion size was measured to determine the effect of FTY720 on de novo atherogenesis. Morphometric analysis in the aortic root did not show any effect of FTY720 on lesion size (10.2x104±1.4x104 µm2 in controls versus 12.5x104± 2.5x104 µm2 in treated group; P=0,66) or fractional area of the lesion in (10.9±1.6% versus 14.3±2.2%; P=0,34) (Figure 4A and 4B and supplemental Figures III and IV). Immunohistochemical analysis of lesion composition yielded no significant differences in T, B cells or macrophage content (supplemental Table II). Expression of cytokine-induced genes (I-Ab, VCAM-1) was not different. RT-PCR analysis of the aorta showed no significant differences in cytokine/mediator pattern (supplemental Figure V).
|
Advanced Atherosclerosis
A second group of ApoE–/– mice on normal laboratory diet with established atherosclerotic lesions (26 weeks old) was treated orally with FTY720 at 3 mg/kg/d for 12 weeks. Similar results with respect to significant lymphopenia and changes in the lipid profile were found as in the early atherosclerosis experiment. All other parameters displayed no difference, comparable with the results in early atherosclerosis (supplemental Table III).
Lesion Size, Cellular Composition, and Cytokine Pattern
Quantitative analysis of the aortic root showed no significant differences between the 2 groups. Lesion size was similar comparing controls with treated animals (43.6x104 ± 4.0x104 µm2 versus 38.5x104 ± 2.7x104 µm2, P=0.15) and also the fractional area of the lesion (24.7±2.1% versus 23.6±1.2%, P=0.26). Immunohistochemical analysis of the advanced lesions detected no differences in T, B cell or macrophage content comparing treated animals with controls (data not shown). To evaluate whether FTY720 had any effect on plaque stability in advanced atherosclerosis we examined the innominate artery.21 Lesion size, fractional area of the lesion, plaque thickness, thickness of the fibrous cap, amount of calcification, and intraplaque hemorrhage were not different between the 2 groups (data not shown). RT-PCR analysis of the aorta showed no significant differences in cytokine/mediator pattern (supplemental Figure VI).
| Discussion |
|---|
|
|
|---|
Three aspects appear noteworthy as to why administration of FTY720 has not been associated with hypercholesterolemia up to present. First, with respect to method only very few groups have performed a detailed plasma lipoprotein analysis as in our study. All currently analyzed clinical trials have not reported cholesterol levels and thus may have missed such an effect.10,11 Second, the choice of animal model may influence the detected effects. In this respect, the ApoE–/– mouse presents a unique model—by means of its disturbed lipid clearance resulting in a massive elevation of plasma cholesterol levels primarily attributable to an increase in cholesterol-rich VLDL—and chylomicron remnant particles. ApoE–/– (and also LDLr–/–) mice are sensitive to a high-fat diet, which leads to a marked increase in non-HDL cholesterol levels.33 It is therefore likely that the effect of FTY720 on plasma cholesterol levels as observed in our study is masked on administration of a high-fat diet. This could explain why this effect was not observed in 2 recent studies on FTY720 in atherosclerosis in hyperlipidemic mouse models.15,16 Third, the administered dose and route will greatly impact on drug levels and effect. The dose (3 mg/kg/body weight) administered in our study translated into drug levels of 3.1 ng/mL which is in the accepted therapeutic range of 1 to 5 ng/mL.34
Despite drug-induced hypercholesterolemia, lesions were not larger in the treated ApoE–/– mice. This suggests a separate attenuating effect on atherosclerosis of FTY720 mediated by immunomodulation. Corroborating a previous study in C57BL/6J mice, we found a diminished peripheral lymphocyte cell pool, which the authors from that study35 attributed predominantly to a decreased release of naïve lymphocytes from the thymus on long-term treatment with FTY720. Future work needs to delineate how long-term treatment differs from short-term treatment to explain this generalized peripheral lymphopenia. Two options are conceivable—either preserved inhibition of thymic lymphocyte egress36 attributable to differential chemokine requirements in distinct lymphoid compartments37 with long-term administration of FTY720, or peripheral depletion via ie, apoptosis.38 Interestingly, we found maintained numbers of Treg and effector memory T cells in lymphoid tissues which argues against a "conventional" lymphodepletion and indicates a functional expansion of effector cells within an overall diminished peripheral lymphocyte pool. Thus, the ability to mount a systemic immune response was not disabled as evidenced by unaltered atherosclerosis-related OxLDL antibody profiles and splenic T cell proliferation. As to the source of Treg and effector memory cells, 2 options are conceivable. Either a thymic-derived natural Treg pool may be constantly self-regenerating in the periphery, or both CD4+ T cell subtypes are continuously regenerated in the periphery after antigen-induced activation. Our finding that effector memory T cell numbers were preserved indicates a peripheral source of at least the memory T cells as they are not derived from the thymus but rather are the result of peripheral antigen activation and subsequent continuous self-renewal. Recent data indicate that peripheral Treg may originate from memory T cells.39 This may help explain our finding of preserved Treg counts and effector memory T cells as Treg would be regenerated from the peripheral memory T cell pool. In support of our findings, previous data show that FTY720 mediated sequestration of effector memory T cells into lymph nodes8 and promoted accumulation of natural regulatory T cells,40 the latter exerting protective effects on atherosclerosis.41 However, lymphocyte retrafficking did not translate into a protective effect on atherosclerosis in our study which is in contrast to 2 very recent studies.15,16 Three reasons may account for the discrepancy between these studies and the present one. First, in both studies animals were fed a high-fat diet which in itself causes a pronounced monocytosis and accumulation of macrophages and T cells in lesions.17–19 In addition, hypercholesterolemia strongly promotes lymphocyte and macrophage activation.42 Thus, the suppressive effect of FTY720 on inflammatory cell trafficking5,12 as well as lymphocyte activation15 may be facilitated under such conditions. Of note, inhibitory effects of FTY720 on splenic T cell proliferation were only observed in LDLr–/– mice exposed to high-fat diet but not when fed a normal laboratory diet (Nofer et al, unpublished results). We found a further increase in S1P plasma levels on FTY720 administration to high-fat fed mice when compared with control animals and FTY720-treated mice on normal diet (data not shown). This might translate into an enhanced effect of S1P on atherosclerosis-related effects (ie, eNOS-induced vasorelaxation6 and reduced lymphocyte activation) in high-fat fed mice. Our data are supported by the recent finding that FTY720 inhibits sphingosine-1-phosphate lyase, the enzyme responsible for S1P degradation.43 Normolipidemic C57BL/6J mice had significantly lower sphingolipid levels compared with hypercholesterolemic ApoE–/– mice.44 Second, the treatment period in both studies was extended (16 and 20 weeks, respectively) enabling detection of even small protective effects on atherosclerosis. Third, the drug level was nearly 25-fold higher compared with our study (68 ng/mL versus 3 ng/mL) in 1 group which showed a decrease in lesion size in the aorta in LDLr–/– mice. Interestingly, in that study a second treatment group with a drug level more in line with levels obtained in our study, no protective effect on atherosclerosis could be detected in the aortic root.15 In the other study drug levels were not measured.16
Our study illustrates a link between sphingolipid and cholesterol metabolism and extends previous data.30,43–45 We show that the sphingosine-1-phosphate (S1P) analogue FTY720 mediates an increase in S1P levels which is associated with increased hepatic HMG-CoA reductase gene expression leading to increased serum cholesterol levels. Further evidence for an interaction between sphingolipid and cholesterol metabolism comes from a recent study which showed that statins—HMG-CoA reductase inhibitors—induce endothelial S1P receptors and mediate vasorelaxation by enhanced eNOS production.46 To establish the effect of sphingosine-1-phosphate agonists such as FTY720 on atherogenesis, it seems preferable to avoid extreme conditions such as the administration of high-fat diets to genetically hyperlipidemic animals. This will help extrapolate data derived from ongoing clinical trials. From the currently available data it appears safe to assume that FTY720 neither attenuates nor increases atherosclerosis in ApoE–/– mice.
| Acknowledgments |
|---|
Sources of Funding
Grants supporting this work were from Novartis (Germany) and Deutsche Forschungsgemeinschaft (KL1398/2-1) to R.K. and T.J.D. (DE591/5-5/5-6), ADUMED Medical Research Foundation to J.R.N., Swedish Research Council, the Grönberg, the Novo Nordisk and the Swedish Heart-Lung Foundations, the Stockholm County Council (ALF) and the Karolinska Hospital to M.R. and G.K.H., respectively.
Disclosures
None.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
increases the expression and activity of microsomal triglyceride transfer protein in the liver. J Biol Chem. 2005; 280: 1224–1229.
. J Biol Chem. 2005; 280: 27879–27887.This article has been cited by other articles:
![]() |
J. E. Hughes, S. Srinivasan, K. R. Lynch, R. L. Proia, P. Ferdek, and C. C. Hedrick Sphingosine-1-Phosphate Induces an Antiinflammatory Phenotype in Macrophages Circ. Res., April 25, 2008; 102(8): 950 - 958. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |