A hallmark of atherosclerosis is the accumulation of free (FC) and esterified (EC) cholesterol in macrophages and smooth muscular cells transforming them in foam cells 1 . Such accumulation depends on the balance between the uptake of cholesterol-rich lipoprotein through scavenger receptors 2 and the efflux of free cholesterol controled by the transporters ABCA1 and ABCG1 3 and to a lesser extent by SR-B1 (or CLA-1) 4 5 . This accumulation depends also on the intra-cellular metabolism of cholesterol, particularly the balance between its free and esterified forms. CE taken up by cells is hydrolyzed in lysosomes to FC that is then directed to various cell membranes by the protein NPC1 6 . Excessive accumulation of FC has toxic effects on cells 7 and FC must be either by removed through efflux to extra-cellular acceptors or esterified. CE is then stored in lipid droplets and can be removed from cells only after hydrolysis to FC by a cholesterol ester hydrolase, whose nature is still debated 8 9 10 .

Storage of lipids droplets in cells accumulating triacylglycerols (adipocytes, hepatocytes) or EC (steroidogenic cells) is also dependent in part of specific proteins surrounding these droplets and belonging to the PERILIPIN family (previously named PAT family) 11 , particularly perilipin1 (previously perilipin) and perilipin2 (previously adipophilin or ADRP). Perilipin2 is present in all cells storing lipids 12 . Its expression is increased during incubation of macrophages with oxidized LDL 13 14 . It is expressed in foam cells of atherosclerotic plaques 14 . Its overexpression in THP-1 macrophages enhances lipid accumulation 15 whereas its invalidation protects against atherosclerosis 16 . Therefore perilipin2 is clearly involved in atheroma. Perilipin1 has at least three different forms, perilipin1 A, B and C, resulting from alternative splicing of a common premessenger RNA 17 . Perilipin1 A and B are present in adipocytes, the A form being largely predominant. Perilipin1 C is found in steroidogenic cells. In adipocytes, perilipin1 opposes in the basal state hydrolysis of triacylglycerols. β-adrenergic agents phosphorylate perilipins1 on specific serine sites and phosphorylated perilipins 1 allow phosphorylated HSL to hydrolyze TG. Perilipins1 are thus implicated in the regulation of basal and stimulated lipolysis. Perilipin1 A is expressed in macrophages 18 19 20 21 and vascular smooth muscular cells 18 , in arterial wall 18 and is overexpressed in atheroma plaques 18 22 , particularly unstable plaques 22 . Perilipin1 could therefore be implicated in the development of atherosclerosis by controling the hydrolysis of stored EC. The overexpression of perilipin1 in atheroma plaque could favour the accumulation of cholesterol and promote the development of atheroma. However, perilipin1 could also shift the balance between FC and EC toward EC and help to prevent excessive accumulation of FC. Since excess FC is toxic for cells and plays a role in the evolution of plaques toward instability 7 23 , perilipin1 could have on the contrary a protective role. In a first step to clarify this issue and to determine the role, if any, of perilipin1 in atheroma, we investigated the effect of perilipin1 invalidation on the development of atheroma in an experimental model, the Ldlr ^-/- mouse.

Plin1 ^-/- mice were a gift from I Tabas (Columbia University, NY, USA). These mice are from the strain generated and described by Martinez-Botas et al 24 and are on a C57BL/6 background. Ldlr ^-/- mice (C57BL/6 background) were from the Jackson Laboratory (Bar Harbor, ME, USA). Plin1 ^-/- and Ldlr ^-/- mice were crossed to obtain Plin ^+/- Ldlr ^+/- mice, which were intercrossed to obtain Ldlr ^-/- and Plin1 ^-/- Ldlr ^-/- mice. All mice were housed in an animal facility with controlled temperature (21-23°C) and lighting (light on 07:00, light off 19:00) and had free access to food and water. Male mice were used for the experiments. These mice received, starting at 8 weeks of age, an atherogenic diet (38.4% fat, 0.15% cholesterol, U8958 version 52 from SAFE, Augy, France) and were investigated after 10 (8 Ldlr ^-/-, 13 Plin1 ^-/- Ldlr ^-/- mice) or 20 weeks (13 Ldlr ^-/-, 30 Plin1 ^-/- Ldlr ^-/- mice) of atherogenic diet. This study was approved by the Ethical Committee of the University Cl Bernard of Lyon.

Blood pressure and heart rate were measured in 20 weeks old mice (8 Ldlr ^-/-and 16 Plin1 ^-/- Ldlr ^-/- mice) by the tail-cuff method (Visitech 2000 series II) after acclimation to restrain and tail-cuff inflation. For quantification of atheroma mice were anaesthetized with pentobarbital. Blood was collected from inferior vena cava and plasma separated by centrifugation for enzymatic determination (Biomerieux, Lyon, France) of total cholesterol and of triacylglycerols concentrations. Aortas were dissected from aortic root to iliac bifurcation, carefully cleaned form periarterial adipose tissue, pinned on silicon dishes and stained for lipid deposits with Red Sudan IV (en face method). Red Sudan IV positive areas were quantified using Image J software and expressed as the percentage of total aorta area. For histological studies some aortas of 20 weeks atherogenic diet fed mice were rinced with PBS and fixed in phosphate buffered formalin, carefully dissected, dehydrated and embedded in paraffin before sectioning. Haematoxylin-eosin and Verhoeff (elastic fibres) stainings were performed as well as immunocytology for macrophages (anti-Mac3 antibodies), lymphocytes (anti-CD3 antibodies) and smooth muscular cells (anti α-SMA antibodies) 25 . For measurements of mRNA concentrations aortas of 20 weeks atherogenic diet fed mice (5 Ldlr ^-/-and 5 Plin1 ^-/- Ldlr ^-/-) were removed, flushed with cold isotonic saline, carefully cleaned of perivascular adipose tissue and flash frozen in liquid nitrogen before storage at -80°C until analysis. Total RNAs were purified using TRIZOL^R protocol (Invitrogen, Cergy-Pontoise, France) with the addition of a DNase treatment. Concentrations, intergrity and purity were verified. For measurements of individual mRNA levels (IL-6, IL1-β, TNFα, MCP-1, SR-A, ABCA1, ABCG1), total RNA was reverse transcripted using Superscript II (Invitrogen) and random hexamers. Real time PCR was performed in a MyIQ thermal cycler (Biorad, Marnes La Coquette, France) using iQ SYBR green Supermix (Biorad). Samples were run along with dilutions of known amounts of target sequence for quantification of initial cDNA copies. Results were calculated as the target over 18S RNA concentration ratio (ng/μg). Primer sequences are shown in table 1.

Results are shown as individual values or as means ± sem. Comparisons between groups were performed by Student t test for non paired values or by Mann-Whitney test using GraphPad Prism (version 5.03). P < 0.05 was considered as indicating a significant difference.

Plasma cholesterol levels were high in both Plin1 ^-/- Ldlr ^-/- and Ldlr ^-/- mice but did not differ between the two groups of mice (table 2). Plasma triacylglycerols levels were comparable. Systolic blood pressure (111.5 ± 3.1 vs 116.5 ± 6.6 mmHg in Ldlr ^-/- mice) and heart rate (547 ± 25 vs 515 ± 23 b/min) were also comparable. Plin1 ^-/- Ldlr ^-/- had a slightly lower body weight (10 weeks: 25.2 ± 0.4 vs 27.8 ± 0.7g p < 0.01; 20 weeks: 27.6 ± 0.5 vs 30.0 ± 0.9g p < 0.05) and as expected 24 an evident decrease at examination of fat pads volume. Despite these comparable blood pressure and plasma lipids levels and decreased fat mass, Plin1 ^-/- Ldlr ^-/- mice had after 10 weeks of atherogenic diet a trend for increase in atherosclerosis (figure 1) as quantified by the en face method. This increase was significant (+55%, p < 0.01) at 20 weeks. Figure 2 shows a representative sample of the extent of lesion in aortas from Plin1 ^-/- Ldlr ^-/- and Ldlr ^-/- mice. Histological examination of plaques performed on aortas from mice fed since 20 weeks the atherogenic diet showed no differences in structure (elastic lamellae, fibrosis, cellularity) between Plin1 ^-/- Ldlr ^-/- and Ldlr ^-/- mice (figure 3). The abundance of lymphocytes and macrophages estimated by immunocytology (figure 4) was comparable and there was no evidence of increased inflammation in plaques of Plin1 ^-/- Ldlr ^-/- mice. In addition we found no increase in the mRNA levels of MCP-1, IL-6 or IL1-β (data not shown) and only a non significant trend for higher values of TNFα mRNA (1.72 ± 0.37 10^-4 vs 1.13 ± 0.26 10^-5 ng/μg 18S RNA p = 0.10) in aortas from Plin1 ^-/- Ldlr ^-/- mice. The expression of SR-A, implicated in the uptake of modified lipoproteins, and of ABCA1 and ABCG1, controlling the efflux of cholesterol, were increased in aortas of Plin1 ^-/- Ldlr ^-/- mice (respectively 3.59 ± 1.06 10^-1 vs 4.81 ± 1.09 10^-2, 5.85 ± 1.53 10^-3 vs 2.94 ± 1.03 10^-4 and 7.17 ± 2.95 10^-4 vs 1.44 ± 0.45 10^-4 ng/μg 18S RNA, p < 0.05 for all).

The present results support an atheroprotective role of perilipin1 since the extent of atheroma lesions was increased in Ldlr ^-/- mice with perilipin1 ablation. The overall histological appearance of the plaques was unchanged and we observed no increase in macrophages or lymphocytes infiltration, nor any significant increase in the mRNA concentrations of pro-inflammatory cytokines and chemokines. Therefore inflammation does not appear increased and the enhanced atheroma seems to result mainly form an increase in lipid deposits. Such an increase is demonstrated by the examination of aortas by the en face method since Red Sudan IV stains lipid deposits. An increased expression of SR-A could contribute to this increase in lipid deposits. The increase in ABCA1 and ABCG1 expression would on the contrary favour cholesterol efflux and could be an adaptative response to the increased accumulation of intra-cellular cholesterol. Perilipin1 can be found in macrophages and vascular smooth muscular cells and is present within atheroma plaques in foam cells originating from both cell types 18 19 20 21 22 . We used mice with a global invalidation of perilipin1 and cannot delineate the respective roles of macrophages and smooth muscular cells perilipin1 in the evolution of atheroma. This will need investigating mice with tissue specific invalidation. We cannot exclude an indirect effect on atheroma of perilipin1 ablation outside of the vasculature, i.e. in adipose tissue. However, blood pressure and plasma lipid levels were unaffected while body weight and fat mass (as appreciated by fat pads volume) were reduced by perilipin1 ablation. Therefore we can exclude a role for three major risk factors for atheroma (hypertension, higher plasma lipid levels, obesity) in the enhanced atherosclerosis we observed. Physical activity was also unchanged (data not shown) in agreement with previous reports 26 . Perilipin1 invalidation induces however some peripheral insulin-resistance, despite the reduced adiposity 27 . We cannot exclude a role for such insulin-resistance.

In conclusion we found a protective role of perilipin1 against atherosclerosis. Evaluation of the respective roles of perilipin1 in macrophages and in smooth muscular cells and elucidation of the precise mechanisms implicated will require further studies.

The authors declare that they have no competing interests.

DL had in charge the mice strains and performed with JYL the quantification of atheroma and histological studies. FF and PDC had in charge mice strains and collected data, particularly plasma lipids and blood pressure. SN had in charge mRNA determinations. MB designed the study and wrote the manuscript. All authors participated in the analysis of data and the preparation of the manuscript.