LRP1 Controls cPLA2 Phosphorylation, ABCA1 Expression and Cellular Cholesterol Export

Background ATP-binding cassette transporter A1 mediates apolipoprotein AI-dependent efflux of cholesterol and thereby removes cholesterol from peripheral tissues. ABCA1 expression is tightly regulated and deficiency of this cholesterol transporter results in cholesterol accumulation within cells. Low-density lipoprotein receptor-related protein 1 (LRP1) participates in lipid metabolism and energy homeostasis by endocytosis of apolipoprotein E-containing lipoproteins and modulation of cellular proliferation signals. Methods and Principal Findings In the present study, we demonstrate a new role for LRP1 in reverse cholesterol transport. Absence of LRP1 expression results in increased PDGFRβ signaling and sequential activation of the mitogen-activated protein kinase signaling pathway, which increases phosphorylation of cytosolic phospholipase A2 (cPLA2). Phosphorylated and activated cPLA2 releases arachidonic acid from the phospholipid pool. Overproduction of arachidonic acid suppresses the activation of LXR/RXR heterodimers bound to the promoter of LXR regulated genes such as ABCA1, resulting in greatly reduced ABCA1 expression. Conclusions and Significance LRP1 regulates LXR-mediated gene transcription and participates in reverse cholesterol transport by controlling cPLA2 activation and ABCA1 expression. LRP1 thus functions as a physiological integrator of cellular lipid homeostasis with signals that regulate cellular proliferation and vascular wall integrity.


Introduction
Cholesterol is an essential component of cell membrane and necessary for normal cellular function, including cell proliferation [1]. Excess cholesterol accumulation, however, can result in pathological consequences. This is particularly true for cells of the arterial wall, where accumulation of cholesterol initiates atherosclerosis [2,3]. A complex homeostatic network has therefore evolved to modulate cholesterol biosynthesis, transport and excretion. Studies on Tangier disease have revealed an important role of ATP-binding cassette transporter A1 (ABCA1) in cholesterol homeostasis [4,5,6]. As a membrane transporter, ABCA1 facilitates the formation of HDL via apolipoprotein AI (apoAI)-mediated efflux of cholesterol and phospholipids from many tissues [7,8,9]. This constitutes the initial step of reverse cholesterol transport, and ultimately leads to the elimination of cholesterol from the body [10,11,12]. Functional defects in the ABCA1 protein that impair its ability to mediate cellular cholesterol efflux can thus result in deposition of cholesterol within the tissues.
As a member of the LDL receptor (LDLR) family, LDL receptorrelated protein 1 (LRP1) was initially identified as a cellular receptor that endocytoses apolipoprotein E (apoE)-enriched lipoproteins [13,14,15,16]. Subsequent studies have shown, however, that LRP1 is a highly multifunctional receptor that not only mediates the endocytosis of a broad spectrum of macromolecules, but also functions as a modulator and integrator of several fundamental cell signaling pathways [17,18,19,20]. One of these involves signaling by platelet-derived growth factor BB (PDGF-BB).
LRP1 forms a complex with the PDGF receptor b (PDGFRb) in clathrin-coated pits and caveolae [17,21,22]. Absence of LRP1 in vascular smooth muscle cells in the mouse (smLRP12/2) leads to increased PDGFRb expression, greatly accelerated development of atherosclerotic lesions, and prominent accumulation of cholesterol in the vessel wall [18]. LRP1 also regulates Wnt5a signaling during adipocyte differentiation and thereby serves as an endogenous regulator of cellular cholesterol and triglyceride homeostasis [20].
Although LRP1 and ABCA1 therefore both play import ant and distinct roles in cellular cholesterol homeostasis and atherosclerosis, the functional interaction between these two membrane proteins has never been investigated. The accumulation of cholesterol in the vascular wall of smLRP2/2 mice, even in the presence of normal or only moderately increased plasma cholesterol levels, and in particular the massive accumulation that occurs in the absence of the LDL receptor suggested a disruption of cholesterol export from the LRP1-deficient smooth muscle cells as a potential underlying mechanism. In the present study, we have taken advantage of the smLRP12/2 mice to investigate the consequences of LRP1 deficiency for ABCA1 expression and function in vitro and in vivo. Our goal was to investigate if and how LRP1 regulates ABCA1 functional expression and thereby cholesterol efflux in the vascular wall.

Results
SmLRP12/2 mice have increased total cholesterol levels in the aorta LRP1 is a multifunctional endocytic receptor participating in the removal of TG-rich VLDL and chylomicron remnants in the liver [23]. Recent studies have shown that LRP1 is also involved in cholesterol storage and fatty acid synthesis in fibroblasts and adipocytes [20,24]. However, cholesterol levels in the aortas of young smLRP12/2 mice in the absence of atherosclerotic lesions have so far not been investigated. Using gas chromatography and mass spectrometry (GC/MS), we found a significant increase in total cholesterol in the aortas of smLRP12/2 mice (Figure 1 and Figure S1). Only aortas lacking any morphologically discernible atherosclerotic lesions or plaques were analyzed. These increased cholesterol levels are thus not caused by the presence of cholesterol in plaques or advanced lesions, but reflect either increased lipoprotein uptake or the inability of LRP1-deficient smooth muscle cells to export endogenous cholesterol in the presence of an intact endothelium.
Lack of LRP1 expression in SMCs results in reduced ABCA1 protein expression in the aorta Because ABCA1 plays an important role in lipid transport [25], impaired expression of ABCA1 in the aorta could potentially be responsible for the increased total cholesterol and the enhanced sensitivity to atherosclerosis in the smLRP12/2 mice. To examine the expression of ABCA1 protein in wild type and smLRP12/2 mice, we immunoblotted and immunostained the aorta with a specific monoclonal rat anti-mouse ABCA1 antibody. As shown in Figure 2A, a significant reduction in ABCA1 protein in the smLRP12/2 mice was revealed by Western blotting. This was supported by immunohistochemical staining of aorta sections, which also showed greatly diminished ABCA1 protein in the smLRP12/2 aortas ( Figure 2B).

Increased cellular lipid accumulation is detected in LRP1deficient SMCs
To characterize the changes of cellular lipids in the presence and absence of LRP1, primary SMCs were generated from the mouse aorta. First, we confirmed by immunoblotting that ABCA1 protein levels are also significantly reduced in primary SMCs lacking LRP1 expression ( Figure 3A). We further detected an increase in LDLR expression in the absence of LRP1. Next, we stained the cells with Oil Red O to visualize neutral lipid deposits. Oil Red O staining showed increased lipid accumulation in the LRP1-deficient cells ( Figure 3B, right panel). To further analyze the components of increased lipids in the LRP12/2 SMCs, we extracted lipids four hours after cellular uptake of 14 C-labelled oleic acid and separated the lipid extracts by thin layer chromatography (TLC). We found increased amounts of choles- The aorta from the aortic root to the iliac bifurcation was isolated, and the surrounding connective tissue was carefully removed under a dissecting microscope. The aorta was weighed after superficial drying with tissue paper. Cholesterol in the aorta was determined by GC/MS and the amount of total cholesterol (mg) was normalized by the weight of the aorta (g). Significantly increased total cholesterol levels were found in the aorta of the smLRP12/2 mice (black bar). 4-month old female littermates of the indicated genotypes were used for the depicted experiment. LRP flox/flox animals lacking the Sm22Cre transgene are indistinguishable from wild type [18] and were used as control groups. Equivalent results were obtained for male mice and throughout the lifetime of the animals ( Figure S1). Values were presented as mean6S.E.M. *p,0.05 (compared to wild type, n = 3/group). doi:10.1371/journal.pone.0006853.g001 Aortas of the indicated genotypes were homogenized in lysis buffer. 10 mg of protein extracts were subjected to 4-15% SDS-PAGE gel, and the blot was then probed with a rat anti-mouse ABCA1 monoclonal antibody. b-actin was detected to demonstrate equal loading. Three independent experiments were quantitated. ABCA1 protein expression in smLRP-aortas was reduced to 0.3960.04 compared to wild type. B. After fixation, aortas were embedded in paraffin and 5 mm sections were prepared. WT (left) and smLRP12/2 (right) aorta sections were then stained with the rat anti-mouse ABCA1 monoclonal antibody, and detected by Alexa-Fluor 568-conjugated goat anti-rat IgG (red). Images were acquired on a Leica TCS SP confocal microscope. Scale bar: 25 mm. Both assays showed greatly reduced ABCA1 protein expression in the aorta of the smLRP12/2 mice. doi:10.1371/journal.pone.0006853.g002 terol ester and free fatty acids ( Figure 3C) in the LRP1-deficient SMCs. These data suggested that LRP1 could potentially regulate cellular lipid trafficking by controlling ABCA1 protein expression.

LRP1 regulates ABCA1 expression at the transcriptional level
To detect whether the reduced ABCA1 protein levels are caused by decreased gene transcription, we next quantitated the mRNA levels of ABCA1 and ATP-binding cassette transporter G1 (ABCG1), another transporter protein involved in cholesterol efflux, both in WT and LRP12/2 SMCs using real-time PCR. The result showed abundant expression of ABCA1 in primary SMCs, whereas transcription of ABCG1 was almost undetectable ( Figure 4). Moreover, we noted a reduction of approximately 80% in ABCA1 mRNA expression in the LRP1-deficient SMCs. This robust decrease of ABCA1 mRNA is consistent with the significant reduction of ABCA1 protein, suggesting that in the absence of LRP1, ABCA1 expression is repressed at the transcriptional level.

Transcription of LXRs and RXRs is not suppressed in the LRP-deficient SMCs
Liver X receptors (LXRs) and retinoid X receptors (RXRs) are the key transcriptional regulators of ABCA1 [25,26]. LXRs form obligate heterodimers with the RXRs and the heterodimers bind to the LXR-responsive elements (LXREs) in the proximal promoter region of ABCA1. Thus, decreased LXR and RXR gene transcription could be responsible for the reduction in ABCA1 mRNA levels. To examine the mRNA levels of LXRs and RXRs in the WT and LRP12/2 SMCs, we performed real-time PCR. Our results showed that the LXR and RXR mRNA levels in the LRP1-deficient SMCs were comparable to those present in the wild type cells and LXRa levels were even increased 2.4 fold ( Figure 5), suggesting that the reduced ABCA1 expression in the LRP12/2 SMCs is not due to changes in gene transcription of LXRs and RXRs.

Repressed ABCA1 transcription in the absence of LRP1 is reversed by LXR agonists
Like most other nuclear receptors that form heterodimers with RXRs, LXRs reside within the nucleus, bound to LXR response elements (LXREs) and complexed with corepressors. In the absence of ligand activation, these corepressors diminish the Figure 3. Lipid analysis of primary SMCs. A. Primary SMCs of the indicated genotypes were lysed in freshly prepared lysis buffer. 10 mg of protein extracts were loaded on 4-15% SDS-PAGE gels. Blots were then probed with the rat anti-mouse ABCA1 monoclonal antibody, a rabbit anti-LDLR polyclonal antibody, and a rabbit anti-LRP1 polyclonal antibody. b-actin was detected to demonstrate equal loading. Western blotting consistently confirmed significantly reduced (,5 fold) ABCA1 and increased LDLR protein expression (1.6760.12, n = 3 and [36]) in primary SMCs lacking LRP1. B. Cells grown on coverslips were fixed with 10% formalin and stained with 4 mg/ml Oil Red O in isopropyl alcohol. Oil red O staining showed increased lipid accumulation in LRP1deficient cells (right panel). Images were captured on a Zeiss fluorescence microscope. Scale bar: 12.5 mm. C. Primary SMCs of the indicated genotypes were harvested and lysed after incubation with 14 C-labelled oleic acid. Lipids were extracted with chloroform/methanol (2:1, v:v) and processed for thin layer chromatography (TLC) analysis. The TLC plates were exposed to a phosphor imaging plate. Cholesteryl ester, triglycerides and free fatty acids in the lipid extracts were separated in a solvent system consisting of hexane/ethyl ether/acetic acid (80:20:1, v:v:v). 3   transcriptional activity of LXRs [27]. To explore whether the LXR agonist T0901317 [28] could activate the LXR/RXR complex, we added different amounts of T0901317 to the LRP12/2 SMCs and performed real-time PCR to follow the transcriptional regulation of ABCA1. SREBP-1c, another LXR target gene, was monitored for comparison. LXRa, LXRb and cyclophilin are not regulated by LXRs and served as controls. As shown in the Figure 6B, T0901317 potently up-regulated ABCA1 and SREBP-1c transcription already at a low concentration of 0.1 mM in the LRP1-deficient SMCs and reached its maximal effect at about 1 mM. A similar dose-dependent response to T0901317 was seen in the WT SMCs ( Figure 6A). However, the relative increase of ABCA1 mRNA expression was several fold greater in the LRP12/2 SMCs than in the wild type cells. Treatment with the LXR activator restored ABCA1 mRNA expression to wild type levels ( Figure 6C). These data suggest that excessive repression of the LXR/RXR complex on the ABCA1 promoter, which can be overcome by treatment with an LXR agonist, may be responsible for its reduced transcription in cells lacking LRP1.
Because ABCA1 regulates the efflux of cellular cholesterol through direct interaction with ApoAI [29,30], we measured ApoAI-mediated cholesterol efflux in the WT and LRP12/2 SMCs in the presence and absence of T0901317. As we expected, a diminished cholesterol efflux was observed in the absence of LRP1 due to the low level of ABCA1 expression. Cholesterol efflux increased significantly in the LRP-deficient cells following T0901317 stimulation along with increased expression of ABCA1 ( Figure 6D).

ABCA1 transcription is suppressed by LXR antagonists
Arachidonic acid can competitively antagonize T0901317dependent activation of the LXR [31]. To examine its effects on the transcription of the LXR target genes ABCA1 and SREBP-1c, we administrated different amounts of arachidonic acid to wild type SMCs and analyzed ABCA1 transcription by real-time PCR. Our data show that arachidonic acid strongly suppresses ABCA1 transcription already at low concentrations of 3 mM and reaches its maximal inhibitory effect at 100 mM (Figure 7). Because the LRP1-deficient cells accumulate more cholesteryl esters and free fatty acids, this result suggest that excessive production of unsaturated fatty acids could be responsible for the inhibition of the LXR-dependent transcription in the absence of LRP1, leading to the observed reduction in ABCA1 expression. The increase in LXRa mRNA expression ( Figure 5) may reflect a compensatory response by the cell to counter this increased inhibition.
Loss of LRP1 function leads to cPLA 2 hyperphosphorylation Extracellular regulated kinase (ERK) 1/2, also called mitogenactivated protein kinase (MAPK), has been reported to phosphorylate cytosolic phospholipase A2 (cPLA 2 ) on Ser 505 , which induces the activation of this 85 kDa cellular enzyme [32,33,34]. Activated cPLA 2 then translocates to membrane vesicles and selectively releases arachidonic acid from the sn-2 position of membrane phospholipids [35]. Thus, increased cPLA 2 activity results in more arachidonic acid production, and this could lead to excessive inhibition of LXR and reduced ABCA1 transcription. To investigate this possibility, we performed Western blotting using a specific anti-phospho-cPLA 2 (Ser 505 ) antibody. In the absence of LRP1, ERK is activated as a result of increased mitogenic, PDGFR-b mediated signaling [17,18]. Significantly increased phosphorylation of cPLA 2 was seen in the absence of LRP1 in response to increased ERK1/2 phosphorylation ( Figure 8A), suggesting that excessive production of arachidonic acid may indeed be responsible for the reduced expression of ABCA1 in the absence of LRP1. To further test this hypothesis, we applied a cPLA 2a inhibitor to the LRP1-deficient SMCs and detected the expression of ABCA1 protein by Western blotting. Consistent with our hypothesis, ABCA1 protein levels where increased in a dosedependent manner by this inhibitor ( Figure 8B). This result shows that LRP1 controls cPLA 2 activity, which in turn regulates the expression of ABCA1.

Discussion
In the present study we have shown that LRP1 controls the expression of ABCA1, a major lipid transport protein in the plasma membrane that mediates cellular cholesterol export. Loss of LRP1 results in greatly reduced ABCA1 protein expression, which is caused by decreased LXR-mediated gene transcription. Increased mitogenic signaling in the absence of LRP1 activates ERK1/2 [17,18], which in turn stimulates cPLA 2 to release arachidonic acid, a potent LXR antagonist, from cellular phospholipids. This potent transcriptional repression of ABCA1 occurs in cultured cells (Figure 3) as well as in the vascular wall in vivo (Figure 2), where it results in a significant reduction of cholesterol export from LRP1-deficient smooth muscle cells ( Figure 6D) and in the accumulation of excess cholesterol in the aortic wall (Figure 1). Reduced ABCA1 expression in the aorta of smLRP12/2 mice thus results in impaired cholesterol efflux and contributes to the excessive cholesterol accumulation in vivo. This mechanism may explain at least in part the greatly accelerated atherosclerosis that occurs in mice lacking LRP1 specifically in their smooth muscle cells. LRP1 thus functions as a physiological integrator of cellular lipid homeostasis with signals that regulate cellular proliferation and vascular wall integrity.
We observed increased accumulation of cholesterol and neutral lipids in the aortic wall and in LRP1-deficient smooth muscle cells (Figure 1 and 3). However, this cannot be explained entirely by increased lipid uptake from the circulation. First, LRP1 is a powerful lipoprotein uptake receptor, and loss of LRP1 expression  thus should result in reduced, not increased cholesterol accumulation. On the other hand, we detected a minor compensatory increase in the level of LDLR expression (,1.7 fold), consistent with earlier observations in the LRP1-deficient liver [36], which may contribute to the cholesterol elevation in LRP1 deficient aortas of LDLR-expressing mice. However, neither increased LDLR-mediated cholesterol uptake nor endogenous cholesterol biosynthesis can explain the massively increased cholesterol accumulation and atherosclerosis that occurs in smLRP-mice that also lack functional LDL receptors [18]. Circulating plasma lipoprotein levels are not altered by the presence or absence of LRP1 in SMCs [18], but are primarily determined by the presence or absence of the LDL receptor in the liver. Taken together, these findings suggested a functional defect in cellular cholesterol export from vascular SMCs as a potential underlying mechanism. We therefore investigated whether the expression of ABCA1, a major cholesterol transporter in the plasma membrane, might be altered in the absence of LRP1. As we had suspected, ABCA1 expression was greatly reduced in LRP1-deficient vessels as well as cultured primary smooth muscle cells. That ABCA1 does indeed have a significant atheroprotective role in vivo is further supported by a recent study from the Hayden laboratory [37], which reported tissue-specific roles of ABCA1 that influence atherosclerosis susceptibility.
What is the mechanism by which LRP1 controls cellular ABCA1 expression and cholesterol export? ABCA1 expression is tightly regulated at the transcriptional level, although post-transcriptional modulation has also been described [12,25,38,39]. In this study, we have shown that ABCA1 is down-regulated at the transcriptional level in the absence of LRP1. Although the most important regulators of ABCA1 expression are the nuclear hormone receptors LXR and RXR [40,41,42] which form obligate heterodimers bound to LXREs within its promoter [26,27], changes in the expression levels of LXRs and RXRs cannot explain the diminished ABCA1 expression in the absence of LRP1 ( Figure 5), suggesting that the reduced ABCA1 expression is caused by transcriptional repression. LXR/RXR heterodimers reside in a complex with corepressors. In the absence of activators, the transcriptional activity of ABCA1 is repressed [27]. Binding of activators to LXR/RXR heterodimers results in a conformational change and ABCA1 transcription [27]. In Figure 6 we have shown that a synthetic ligand of LXR, T0901317 [28], is able to activate the transcription of ABCA1, restore ABCA1 mRNA to wild type levels, and normalize ApoAI-mediated cholesterol efflux in the LRP1-deficient SMCs. These results suggested that in the absence of LRP1, although cells express normal amounts of LXRs and RXRs, the LXR/RXR heterodimers are bound to their co-repressors and are thus functionally repressed. This explains the greatly diminished basal expression of ABCA1 in the LRP12/2 cells. In the presence of the synthetic ligand, the LXR/RXR heterodimers are activated and readily induce ABCA1 transcription. The much lower baseline expression of ABCA1 in the LRP1-deficient SMCs further suggested, that an endogenous LXR antagonist may be responsible for this nearly complete transcriptional repression.
TLC analysis of total cellular lipid extracts from wild type and LRP12/2 cells showed that LRP1-deficient SMCs accumulate  more free fatty acids compared to wild type cells. Polyunsaturated fatty acids are known to competitively antagonize activation of LXR by oxysterols and T0901317 [31]. To explore if an excess of polyunsaturated fatty acids could mediate the repression of LXRdependent ABCA1 transcription in the absence of LRP1, we incubated wild type SMCs with increasing concentrations of arachidonic acid. The powerful reduction of ABCA1 and SREBP-1c transcription in the wild type cells by administration of exogenous arachidonic acid suggests that excessive accumulation of polyunsaturated fatty acids in the LRP1-deficient cells could indeed explain the observed reduction of ABCA1 expression in these cells.
As a polyunsaturated fatty acid, arachidonic acid is released from the sn-2 position of phospholipids by phospholipase A2 (PLA 2 ). Mammalian cells contain several forms of PLA 2 including secretory PLA 2 (sPLA 2 ), calcium-independent PLA 2 , and a cytosolic cPLA 2 . cPLA 2 shares no homology with other PLA 2 enzymes, and is the only well characterized PLA 2 that preferentially hydrolyzes arachidonic acid from phospholipids at the sn-2 [34,43]. In addition to being converted to potent inflammatory lipid mediators, which may independently contribute to the increased atherosclerosis of smLRP2/2 mice by promoting macrophage recruitment, arachidonic acid is itself a key regulator of cellular signaling. The importance of arachidonic acid thus ensures that its levels are tightly controlled. As the crucial enzyme in mediating arachidonic acid release, cPLA 2 is rapidly activated by increased concentrations of cytosolic Ca 2+ and by serine phosphorylation [44]. cPLA 2 contains a consensus sequence (Pro-Leu-Ser 505 -Pro) for phosphorylation by the MAPK ERK. ERK efficiently phosphorylates cPLA 2 at Ser 505 , which increases its enzymatic activity [32]. Previous studies have shown that loss of LRP1 expression results in the elevated expression of PDGFRb, which subsequently activates downstream ERK signaling [17,18,45]. Consistent with our previous findings [18], we observed increased phosphorylation of ERK1/2 in the absence of LRP1 (Figure 8), resulting in increased phosphorylation and activation of cPLA 2 , and thus accelerated phospholipid and arachidonic acid turnover. Administration of a cPLA 2 inhibitor increased the expression of ABCA1 in the LRP-deficient cells, which further supports a mechanism by which over-production of polyunsaturated fatty acids, including arachidonic acid, is probably the underlying cause for the reduced ABCA1 expression in the absence of LRP1. Thus, suppression of ABCA1 transcription in the LRP1-deficient SMCs is likely due to repression of the LXR/RXR complexes by an endogenously produced antagonist and this is just one example for the altered LXR activity mediated by increased cPLA 2 activity.
In summary, the present study has shown that LRP1 participates in apoAI-mediated efflux of cholesterol and phospholipids by controlling ABCA1 expression at the transcriptional level. These findings further emphasize the importance of LRP1 as an integrator of lipoprotein transport and cellular signals that regulate cell proliferation and migration, as well as cellular lipid homeostasis.

Animals
WT and smLRP12/2 mice were maintained on a 129SvEv and C57BL/6J hybrid background by intercrossing of Sm22Cre+; LRP flox/flox with hybrid LRP flox/flox animals from the larger colony pool to prevent allele fixation. Sex and weight-matched littermates were used for all experiments. All mice were housed in an animal facility with 12h light/12h dark cycles. The animals were fed a standard rodent chow diet (Diet 7001, Harlan Teklad, Madison, WI) and water ad libitum. Male and female animals between 2 and 18 months of age were used throughout the studies. No sexual dimorphism of phenotype was observed. All procedures were performed in accordance with the protocols approved by the Institutional Committee for Use and Care of Laboratory Animals of the University of Texas Southwestern Medical Center at Dallas.

Primary SMC culture
Primary mouse aortic SMCs were generated using the explant technique as previously described [46]. Briefly, aortas were dissected out under sterile conditions and rinsed twice with PBS. The connective tissue and adventitia were carefully removed. The aorta was opened longitudinally and the intima on the luminal surface was scraped off. The aorta was then cut into small pieces and transferred into a T25 flask containing high glucose (4.5 g/L) DMEM supplemented with 15% fetal calf serum (FCS), 100 U/ml penicillin and 100 mg/ml streptomycin. Outgrowing smooth muscle cells were detached by incubation with 0.25% trypsin-EDTA solution and cultured at 37uC in 5% CO 2 .

Analysis of cholesterol levels in mouse aorta
Sterols in the mouse aorta were analyzed by gas chromatography and mass spectrometry (GC/MS) as previously described [47]. Briefly, the entire aorta from the aortic root to the iliac bifurcation was dissected out and the adventitia was removed. The aorta was weighed after superficial drying with tissue paper. Aortas were then dissolved in 1 ml of 0.1 M NaOH and vortexed for 30 min. An aliquot of ethanol containing the internal standards 5a-cholestane (50 mg) and epicoprostanol (2 mg) was added to 100 ml of tissue lysate, and sterols were hydrolyzed by heating to 100uC in 100 mM ethanolic KOH for 2 h. Lipids were extracted in petroleum ether, dried under nitrogen, and derivatized with hexamethyldisilazanetrimethylchlorosilane. GC/MS analysis was performed by using a 6890N gas chromatograph coupled to a 5973 mass selective detector (Agilent Technologies, Palo Alto, CA). The trimethylsilylderived sterols were separated on an HP-5MS 5%-phenyl methyl polysiloxane capillary column (30 m60.25 mm inner diameter x 0.25 mm film) with carrier gas helium at the rate of 1 ml/min. The temperature program was 150uC for 2 min, followed by increasing the temperature by 20uC per min up to 280uC and holding it for 13 min. The injector was operated in the splitless mode and was kept at 280uC. The mass spectrometer was operated in selective ion

Immunofluorescence microscopy
Mice were perfusion-fixed via the left cardiac ventricle with warmed Hank's balanced salt solution-Hepes (20 mM, pH 7.3), followed by the same solution containing 4% (w/v) paraformaldehyde. The aorta was removed and divided into small pieces. Tissues were immersion-fixed for an additional hour followed by treatment with a mixture of 60% (v/v) methanol, 10% (v/v) glacial acetic acid, 30% (v/v) inhibisol (1,1,1-trichloroethane) for 24 h. Tissues were then embedded in paraffin. Paraffin sections (5 mm thick) were dewaxed in three changes of xylene (10 min each), and rehydrated into PBS (10 mM phosphate buffer, pH 7.2, 0.15 M NaCl). The sections were washed with 50 mM NH 4 Cl in PBS for 30 min and blocked by incubation for 1 h with TBS (10 mM Tris-HCl, pH 9.0, 150 mM NaCl) containing 10% (v/v) normal goat serum and 1% (w/v) bovine serum albumin (BSA). Samples were then incubated overnight at 4uC with a rat monoclonal antibody against mouse ABCA1 (NB400-164, Novus) at a 1:20 dilution. Sections were washed three times in TBS containing 0.1% BSA, and the bound primary antibody was detected by incubation for 2 h with Alexa-Fluor 568-conjugated goat anti-rat IgG (10 mg/ml, Molecular Probes, Eugene, OR). The tissue slides were then washed three times in TBS containing 0.1% BSA, rinsed with water, and mounted on a coverslip with Fluorescence Mounting Medium (DakoCytomation). Images were taken using Leica TCS SP confocal microscope.

Oil Red O staining
Cells were grown on coverslips for two days and then fixed with 10% formalin in phosphate-buffered saline for 1 h at room temperature. After washing three times with deionized water, cells were stained with Oil Red O in isopropyl alcohol at a concentration of 4 mg/ml. Finally, each coverslip was washed for 10 min with deionized water and mounted on glass slides for microscopic evaluation.

Lipid analysis by thin layer chromatography (TLC)
SMCs were harvested and lysed in buffer B (250 mM sucrose, 100 mM KCl, 50 mM Tris-HCl, pH 7.4) after 4 h incubation with 14 C-labelled oleic acid. Part of the cell lysate was utilized for protein measurement. Cell lysate containing equal amounts of protein was further processed for lipid extraction by 500 ml of chloroform/ methanol (2:1, v:v). Lipid extracts were then dried and dissolved in 130 ml of chloroform/methanol (1:1, v:v). 60 ml of the lipid extracts were loaded (30 ml/lane, duplicate) onto a 20620 cm silica gel plate (805013, POLYGRAM SIL G, MACHEREY-NAGEL) for cholesterol ester, triglyceride, and free fatty acid separation by TLC. 20 ml of 3 H-labelled cholesterol recovery solution containing cholesterol oleate ester (30 mg), triglyceride (10 mg) and oleate (20 mg) was used as lipid standards. Cholesteryl ester, triglyceride, and free fatty acid separation was performed using hexane/ethyl ether/acetic acid (80:20:1, v:v:v) as the solvent. TLC plates were dried in a hood, and exposed to a phosphor imaging plate for 24 h at room temperature. The imaging plate was analyzed by the Storm 820 Phosphor imager (Molecular Dynamics, Sunnyvale, CA).

Quantitative real-time PCR
Total RNA from SMCs was prepared using RNA STAT-60 from Tel-Test Inc. Equal amounts of RNA were treated with RNAase-free DNAase I (Ambion Inc.). First-strand cDNA was synthesized from 2 mg of DNAase I-treated total RNA with random hexamer primers using TaqMan Reverse Transcription Reagents (N808-0234, Applied Biosystems). Specific primers for each gene were designed using Primer Express software (Applied Biosystems). The real-time PCR reaction was set up in a final volume of 20 ml containing 20 ng/ml cDNA, 2.5 mM forward and reverse primers, and 10 ml of 2x SYBR Green PCR Master Mix (4312704, Applied Biosystems). PCR reactions were carried out in a 384-well plate using the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems). All reactions were done in triplicate. The relative amount of mRNA was calculated using the comparative threshold cycle (DDC T ) method as recommended by the manufacturer in the Applied Biosystems protocols. The primers for real-time PCR were listed as follows: mouse ABCA1, 59-CGTTTCCGGGAAGTGTCCTA-39 (

ApoAI-mediated cholesterol efflux assay
ApoAI-mediated cholesterol efflux was analyzed as previously described [48]. Briefly, SMCs were seeded into 16 mm wells and radiolabeled starting at 60% confluence in high glucose (4.5 g/L) DMEM containing 10% FCS and 0.3 mCi/ml [ 3 H]-cholesterol (C8794, Sigma). Confluent cells were then loaded with 30 mg/ml unlabeled non-lipoprotein cholesterol for 24 h, and equilibrated for 24 h prior to 24 h incubation of 10 mg/ml apoAI. Equilibration and apoAI administration were performed in the presence or absence of 10 mM T0901317. At the end of the incubation, media were collected and centrifuged at 2,000 xg for 10 min to remove cell debris. Radioactivity in the medium was measured by liquid scintillation counting (LS-60001C, Beckman Instruments Inc.). Cellular lipids were extracted and analyzed for [ 3 H]-sterol. This assay was performed in quadruplicate. Data are expressed as percentage of total (cell plus medium) [ 3 H]-sterol appearing in the medium. Values are expressed as mean6S.D.

Statistics
Unpaired two-tail Student t-test was used for statistical analyses. A p value , 0.05 was considered significant. Figure S1 GC/MS analysis of cholesterol in the mouse aorta. The aorta from the aortic root to the iliac bifurcation was isolated, and the surrounding connective tissue was carefully removed under a dissecting microscope. The aorta was weighed after superficial drying with tissue paper. Cholesterol in the aorta was determined by GC/MS and the amount of total cholesterol (mg) was normalized by the weight of the aorta (g). Increased free and esterified cholesterol levels were found in the aorta of 18 month old smLRP12/2 mice (compared to wildtype mice of the same age, n = 4/group). Found at: doi:10.1371/journal.pone.0006853.s001 (0.19 MB TIF)