at SciVerse ScienceDirect

Atherosclerosis 229 (2013) 408e414

Contents lists available

Atherosclerosis

journal homepage: www.elsevier .com/locate/atherosclerosis

EP 80317, a CD36 selective ligand, promotes reverse cholesteroltransport in apolipoprotein E-deficient mice

Kim Bujold a,3, Katia Mellal a,3, Karina F. Zoccal a,1, David Rhainds a,2, Louise Brissette b,Maria Febbraio c, Sylvie Marleau a, Huy Ong a,*

a Faculté de pharmacie, Université de Montréal, QC, CanadabDépartement des sciences biologiques, Université du Québec à Montréal, QC, Canadac Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB Canada

a r t i c l e i n f o

Article history:Received 31 May 2012Received in revised form8 May 2013Accepted 31 May 2013Available online 14 June 2013

Keywords:CD36Reverse cholesterol transportABCG5ABCG8NPC1L1

* Corresponding author. Faculté de pharmacie, Uni6128, Station Centre-Ville, Montréal, Québec H3C 36460; fax: þ1 514 343 2102.

E-mail address: huy.ong@umontreal.ca (H. Ong).1 Faculty of Pharmaceutical Sciences of Ribeirão Pre

Brazil.2 Present address: Montreal Heart Institute, Montr3 K Bujold and K Mellal have equal contribution to

0021-9150/$ e see front matter � 2013 Elsevier Irelahttp://dx.doi.org/10.1016/j.atherosclerosis.2013.05.031

a b s t r a c t

Aims: The CD36 selective ligand, EP 80317, features potent anti-atherosclerotic and hypocholesterolemiceffects that are associated with an increase in macrophage cholesterol efflux through the activation of theperoxisome proliferator-activated receptor geliver X receptor a (LXRa)eATP-binding cassette (ABC)transporter pathway. Cholesterol efflux is the first step of reverse cholesterol transport (RCT). However,whether EP 80317 exerts its hypocholesterolemic and anti-atherosclerotic activity through RCT in vivohas yet to be determined. In the present study, we investigated the effects of EP 80317 on RCT, inparticular on macrophage-to-feces RCT and the expression of selected genes associated with hepaticcholesterol metabolism and intestinal cholesterol transport.Methods and results: Reverse cholesterol transport was assessed following the intraperitoneal injectionof [3H]-cholesterol-labelled J774 macrophages to hypercholesterolemic apoE- and apoE/CD36 double-deficient mice that had been treated for 12 weeks with EP 80317. Forty-eight hours after the adminis-tration of [3H]-cholesterol-labelled cells, blood, liver, intestines and feces were harvested. The radioac-tivity recovered in the feces (cholesterol and bile acid combined) was significantly increased by 311%(P ¼ 0.0259) in EP 80317-treated mice compared with that found in vehicle-treated mice despite nosignificant change in [3H]-tracer recovery in plasma between groups. Whereas the mRNA levels of LXRain the gut were significantly upregulated, mRNA and protein levels of the NiemannePick C1-like 1protein (NPC1L1) transporter, a LXRa target which regulates intestinal cholesterol absorption, weredownregulated in EP 80317-treated mice. In contrast, neither mRNA nor protein levels of investigatedtransporters and receptors were modulated in the small intestine of double-deficient mice, nor was thefecal recovery of radioactivity. No change was observed in targeted genes in liver of either apoE- or apoE/CD36 double-deficient mice after a chronic treatment with EP 80317.Conclusion: This study shows that EP 80317 elicits macrophage-to-feces reverse cholesterol transport in amanner dependent on CD36 expression. This effect is associated with the upregulation of LXRa and thedownregulation of NPC1L1 expression.

� 2013 Elsevier Ireland Ltd. All rights reserved.

versité de Montréal, P.O. BoxJ7, Canada. Tel.: þ1 514 343

to, University of São Paulo, SP,

éal, Québec, Canada.the work.

nd Ltd. All rights reserved.

1. Introduction

The macrophage CD36 receptor has been shown to play aprominent role in scavenging oxidatively modified low density li-poprotein (oxLDL) and generating foam cells [1]. OxLDL catabolismwithin macrophages has been associated with the generation ofbioactive lipids which activate peroxisome proliferator-activatedreceptor g (PPAR-g) leading to the upregulation of a number ofgenes involved in lipid metabolism, including CD36 [2]. Along thatline, our previous studies showed that EP 80317, a synthetic peptideligand of CD36, presents potent anti-atherosclerotic effects associ-ated with reduced vascular inflammation and hypocholesterolemia

K. Bujold et al. / Atherosclerosis 229 (2013) 408e414 409

following a long-term (>6 weeks) treatment at a dose of 300 mg/kgper day in apoE-deficient (apoE�/�) mice [3,4]. At the cellular level,we showed that murine macrophages incubated with EP 80317exhibited increased cholesterol and phospholipid efflux throughPPARg activation [5]. The latter was shown to raise 15d-PGJ2 pro-duction throughanERK1/2-dependent COX-2upregulation, therebypromoting cholesterol and phospholipid efflux from macrophages[5]. In this scheme of events, PPARg signalling in macrophages isexpected to upregulate ATP-binding cassette (ABC) transporters andincrease the effluxof cholesterol to extracellular acceptors along theHDL pathway. This is the first step of a multi-step process calledreverse cholesterol transport (RCT), resulting in the relocation ofcholesterol from the peripheral tissues to the liver, with subsequenthepatobiliary cholesterol secretion and excretion into feces [6e8].Alternatively, cholesterol may be cleared from the periphery by aliver-independent, transintestinal cholesterol excretion (TICE)pathway, implying a direct cholesterol transport from blood to theintestinal lumen [8e10]. Although detailed mechanisms of TICEremain to be elucidated, key players in cholesterol absorption/secretion by the apical membrane of enterocytes lining the smallintestine include NiemannePick C1-like 1 protein (NPC1L1), theABC transporters G5 and G8 and the CD36 receptor, the latter alsoinvolved in chylomicron formation [11].

The aim of the present study was to determine whetherenhancedmacrophage cholesterol efflux associatedwith prolongedadministration of EP 80317 promoted RCT in a CD36-dependentmanner. To this aim, apoE�/� and apoE/CD36 double-deficient(apoE�/�/CD36�/�) mice were fed with a high fat high cholesterol(HFHC) diet from 6 weeks of age and pretreated with EP 80317 bydaily s.c. injection for 6e12 weeks. Mice were then injected intra-peritoneally with [3H]-cholesterol-labelled J774 murine macro-phages and the appearance of [3H]-cholesterol in the plasma, liver,and feces over the next 48 h was quantitated. The expression levelsof key cholesterol transporters in liver and small intestine havebeen assessed.

Our results show that a chronic treatment with EP 80317 pro-motes macrophage-to-feces RCT, an effect associated with theupregulation of genes and proteins involved in intestinal choles-terol transport and a downregulation of those associated with in-testinal cholesterol absorption.

2. Materials and methods

2.1. Materials

[1a, 2a (n)-3H]-Cholesterol (35e50 Ci/mmol) was purchasedfrom Amersham Biosciences. EP 80317 (HaiceD-2MeTrpeD-LyseTrpeD-PheeLysNH2) was provided by Ardana Bioscience (Edin-burgh, UK). All solutions for parenteral administration were sterile.

2.2. Animals

All experimental procedures were approved by the InstitutionalAnimal Ethics Committee of the Université de Montréal, in accor-dancewith the Canadian Council on Animal Care guidelines and theGuide for the Care and Use of Laboratory Animals published by theUS National Institutes of Health (A5213-01). ApoE�/� and apoE�/

�/CD36�/� mice and their littermate controls were backcrossed sixtimes to C57BL/6 mice [4]. Six-week-old male mice were fed anHFHC diet (D12108, cholate-free AIN-76A semi-purified diet,Research Diets Inc.) containing 40% wt/wt fat and 1.25% wt/wtcholesterol and water ad libitum. Mice were treated with EP 80317(300 mg/kg) or vehicle (0.9% NaCl) administered by daily s.c.injection for 12 weeks.

2.3. Plasma lipid analysis

Total plasma cholesterol and HDL cholesterol were assayed us-ing the Infinity� cholesterol reagent from Thermo Fisher Scientificaccording to the manufacturer’s instructions. QUANTOLIP� HDL(HDL2/HDL3) precipitation reagent from Technoclone was used toprecipitate non-HDL cholesterol from plasma.

2.4. Cell culture

J774 murine macrophages were obtained from the AmericanType Culture Collection. J774 macrophages were cultured in Dul-becco’s minimal essential media (DMEM) supplemented with 10%FBS, 100 UI/ml penicillin, and 100 mg/ml streptomycin. J774 cellswere radiolabelled by incubating the cells with 5 mCi/ml of [1a, 2a(n)-3H]-cholesterol for 48 h at 37 �C. The cells were washed threetimes, equilibrated in DMEM containing 0.2% BSA overnight andresuspended in PBS.

2.5. In vivo reverse cholesterol transport

ApoE�/� and apoE�/�/CD36�/�micewere injected i.p. with [3H]-cholesterol-labelled J774 cells (typically 5 � 106 cells containingbetween 3 and 4 � 106 counts per min (cpm) in 0.5 ml PBS). Bloodsamples (50 ml) were collected from the saphenous vein at 2, 6, 24and 48 h and plasma radioactivity was quantitated by liquid scin-tillation counting (Hionic Fluor scintillation fluid, Perkin Elmer)using a beta counter (Perkin Elmer). After 48 h, mice were eutha-nized by CO2 asphyxiation and exsanguinated, perfused with ice-cold PBS and the organs (liver and jejunum) were removed. Ali-quots (100 mg) of liver and jejunum were dissolved in 2 ml ofSolvable� reagent (Perkin Elmer), incubated at 50 �C overnight,and treated with 0.2 ml of 30% hydrogen peroxide. Feces werecollected at 24 and 48 h and weighed. Radioactivity content inplasma, liver, intestine and feces was expressed as percent countsrelative to total injected tracer.

2.6. Real-time PCR analysis of hepatic and intestinal mRNA levels

In additional experiments, total mRNA was extracted from liverand jejunum (100 mg aliquots) in Trizol (Invitrogen Life Technolo-gies) and reverse transcribed to cDNA using oligo dT 12e18 randomprimers and SuperScript� II reverse transcriptase (Invitrogen) ac-cording to the manufacturer’s instructions. Real-time PCR was per-formed using Platinum� SYBR� green qPCR supermix (Invitrogen)and gene-specific oligonucleotides (Supplemental materialsTable 1) with the Rotor-Gene 3000 instrument (Montreal Biotech).mRNA levels were normalized to the housekeeping gene GAPDHand the relative gene expression (vs. the mean Ct of the vehiclegroup) was calculated using the comparative 2�DDCt method.

2.7. Western blot of intestinal NPC1L1 and ABCG8 transporters

Jejunum enterocytes isolated from 6 weeks treated apoE�/� andapoE�/�CD36�/� mice were weighted and resuspended in ice-coldTris buffer containing complete protease inhibitor cocktail (PIC)(50 mM TriseHCl pH 7.4). The samples were homogenized using aPottereElvehjem glass-Teflon homogenizer. Crude membraneisolationwas obtained after centrifugation at 35,000� g for 20min at4 �C. The pellet fraction was then resuspended in RIPA lysis buffer(50 mM TriseHCL pH 7.5, 150 mM NaCl, 1% Triton X-100, 0.1% SDS,25 mM NaF) containing complete PIC and the protein content solu-bilizedby incubationofmembranepellets at4 �C for1h followedbya10 s sonication. Lysates were centrifuged at 17,000� g at 4 �C for 1 h.The supernatants were assayed for protein concentration and

Table 1Plasma lipid profile in apoE�/� fed with an HFHC diet treated with or without EP80317 (300 mg/kg/day) for 12 weeks.

ApoE�/� mice (mmol/l)

0.9% NaCl EP 80317

Total cholesterol 30 � 2 24 � 1a

Non-HDL cholesterol 29 � 2 23 � la

HDL cholesterol 1 � 0.03 1 � 0.01

aP < 0.05 compared to 0.9% NaCl.Results are means � SEM of 16 and 17 mice in vehicle and EP 80317 group.

K. Bujold et al. / Atherosclerosis 229 (2013) 408e414410

analysed by SDS-PAGE on 4e7.5% gradient in running buffer, trans-ferred onto nitrocellulose membranes and immunoblotted withspecific primary antibodies, to NPC1L1 (Novus), ABCG8 (Novus), andb-actin at 4 �C overnight, followed by incubation with horseradishperoxidase-conjugated secondary antibody for 1 h at room temper-ature. Signals were recorded using ECL Plus reagent (AmershamBiosciences) and were analysed with ImageQuant 5.2 software.

2.8. Statistical analysis

Data are expressed as mean � SEM. Comparisons between twogroups were performed using Student’s unpaired two-tailed t-test,having first checked for homogeneity of variances using the F ratiotest. Differences in group means were considered significant whereP < 0.05.

3. Results

3.1. Effect of EP 80317 pretreatment on plasma lipoproteins andcholesterol levels and RCT after intraperitoneal administration ofpre-labelled J774 macrophages

Total and non-HDL plasma cholesterol levels were reduced by20 and 23% (P ¼ 0.0300), respectively, in EP 80317-treated apoE�/�

Fig. 1. Fate of [3H]-cholesterol following the injection of [3H]-cholesterol-loaded J774 macropweeks, were injected i.p. with [3H]-cholesterol-loaded J774 macrophages. Plasmaetime curveinjection of radiolabelled-macrophages (A). Area under curve (AUC) is expressed as (pmol �expressed as [3H]-tracer % of injection/100mg� SEM (n¼ 11 and 13mice in 0.9% NaCl and EP 8

micewhereas therewas no change inmean HDL plasma cholesterollevels (Table 1). The radioactivity remaining in the plasmacompartment and tissues after the injection of [3H]-cholesterol-labelled J774 macrophages was expressed as a fraction of the totalactivity administered. [3H]-tracer recovery in plasma plateaued byw6 h in both EP 80317- and vehicle-treated groups (Fig. 1A), yet nosignificant change in the area under the plasma concentrationetime curves (0e48 h) of the [3H]-tracer could be observed (Fig. 1B).However, [3H]-tracer recovery of EP 80317-treated mice wasreduced by 42% (P ¼ 0.0141) in the liver (Fig. 1C), while it wasincreased by 311% (P ¼ 0.0259) in the feces (Fig. 1D).

3.2. Effect of EP 80317 on hepatic and intestinal gene expression

We investigated changes in mRNA levels of the major proteinsinvolved in cholesterol metabolism and transport in the liver andintestine. In the liver, neither LXRa and ABC transporters mRNAlevels, nor those of the rate-limiting bile acid synthesis enzymecholesterol 7 alpha-hydroxylase (CYP7a), were modulated afterchronic EP 80317-treatment (Fig. 2A). In addition, we did notobserve significant changes in the fractional catabolic rate of radi-oiodinated lipoproteins, including both LDL and very low densitylipoprotein (VLDL), in EP 80317-treated mice (SupplementaryFig. 1). In contrast, jejunal LXRa mRNA levels were increased by2.3-fold (P ¼ 0.0292) in EP 80317-treated mice. The treatment alsoincreased ABCA1, ABCG1, ABCG5 and ABCG8 mRNA levels by 6.5-(P ¼ 0.0080), 6.7- (P ¼ 0.0255), 4.3- (P ¼ 0.0555) and 5.6-fold(P ¼ 0.0401) respectively, compared to vehicle-treated mice(Fig. 2B). Moreover, the intestinal sterol influx transporter Nie-mannePick C1 like 1 (NPC1L1) mRNA levels were decreased by 70%(P ¼ 0.0019) and those of CD36 were unchanged (Fig. 2B).

3.3. Effect of EP 80317 pretreatment on intestinal NPC1L1 andABCG8 protein expression in apoE�/� mice

The expression of NPC1L1 protein in small intestinewas reducedby 41% compared to 0.9% NaCl-treated mice (Fig. 3A), correlating

hages in vivo. ApoE�/� mice, pretreated or not with EP 80317 (300 mg/kg per day) for 12s of [3H]-tracer (0e48 h), expressed as [3H]-tracer % of injection/ml plasma, following theh/ml) (B). Radioactivity content of liver at 48 h (C) and of feces (0e48 h) (D). Values are0317 groups, respectively). *P< 0.05 comparedwith the vehicle (0.9%NaCl)-treated group.

Fig. 2. Hepatic and intestinal mRNA levels of genes involved in the metabolism andtransport of cholesterol as determined by real-time RT-PCR. Tissues were obtainedfrom apoE�/� mice treated or not with EP 80317 (300 mg/kg per day) for 12 weeks.Total RNA was extracted from liver (A) and jejunum (B) and mRNA levels of indicatedgenes were determined. Data are presented as fold change (�SEM) versus vehicle.Numbers of mice were 11e14 (livers) and 13e17 (intestines). *P < 0.05 and **P < 0.01compared with vehicle (0.9% NaCl)-treated mice.

K. Bujold et al. / Atherosclerosis 229 (2013) 408e414 411

with reduced mRNA levels (Fig. 2B) following long-term (6 weeks)treatment with EP 80317. In contrast, no significant change inABCG8 protein expression was observed (Fig. 3B).

3.4. Effect of EP 80317 pretreatment in apoE/CD36 double-deficientmice

The plasma concentrationetime profile of [3H]-tracer in plasmaof apoE�/�/CD36�/� mice was similar to that observed in apoE�/�

mice following the injection of radiolabelled macrophages, with nosignificant effect of EP 80317 pretreatment on [3H]-cholesterolplasma levels and the area under the plasma concentrationetimecurves (0e48 h) (Fig. 4A and B). Furthermore, EP 80317 did notsignificantly alter hepatic or fecal [3H]-tracer recovery (Fig. 4C andD). In contrast to apoE�/� mice, neither jejunal mRNA levels ofinvestigated genes nor NPC1L1 and ABCG8 protein levels werealtered in mice pretreated with EP 80317 (Fig. 4E and F).

4. Discussion

The principal finding of the present study is that the anti-atherosclerotic GHRP derivative, EP 80317, promotes macrophage-to-feces reverse cholesterol transport in apoE�/� mice in a CD36-dependent manner. Enhanced cholesterol excretion into feces

was mainly associated with the downregulation of intestinalNPC1L1, a known LXR target gene [12]. Although the hepatobiliaryroute has been proposed as the main route for the disposal ofcholesterol [6e8], several recent studies provided evidence foralternative routes for the removal of excess cholesterol from theperiphery [9,10,13,14]. In agreement, Mdr2�/� and ABCG5�/

�/ABCG8�/� deficient mice have diminished ability to secrete bileinto the intestine, yet have similar fecal cholesterol outputcompared with their wild type littermates [15,16]. These observa-tions suggested that macrophage RCT may occur in the absence ofbiliary cholesterol secretion through a pathway now referred to asTICE [8,17].

In our previous studies, we have shown that the anti-atherosclerotic effect observed after a prolonged (>6 weeks)administration of EP 80317 was associated with a reduction in totaland non-HDL-cholesterol in apoE�/� mice fed an HFHC diet [3]. Inthese studies, we showed that incubation of macrophages withsynthetic selective ligands of CD36 upregulated proteins involvedin the reverse transport of cholesterol and increased PPARg activitywithin macrophages [5]. The molecular mechanisms triggeringPPARg activation and enhancing cholesterol efflux involve an ERK1/2-dependent COX-2 stimulation of 15d-PGJ2, an endogenous acti-vator of PPARg [5]. Interestingly, the activation of PPARgeLXRaeABC transporter pathway by EP 80317 was uncoupled from theregulatory positive feedback on CD36 expression observed duringoxLDL uptake, pointing to an interesting selectivity of the phar-macologic response towards the efflux pathways. Yet, whether EP80317 may regulate cholesterol transporters expression at the in-testinal/hepatic levels, and whether macrophage cholesterol effluxcontributed to RCT and to the hypocholesterolemic effect of EP80317 were unknown.

To study whether EP 80317-mediated increased cholesteroleffluxwas coupled to RCT in vivo, we have traced the fate of tritiatedcholesterol following an i.p. injection of [3H]-cholesterol-J774macrophages, according to Zhang et al. [18]. Noteworthy, as in theapoE�/� macrophages, J774 murine macrophages do not expressapoE [19]. In agreement with a preponderant role for intestinal LXRactivation in stimulating reverse cholesterol transport [20], EP80317-elicited RCT was associated with an increase in LXR mRNAlevels in jejunum (Fig. 2B). Among LXR target genes, ABCG5 andABCG8 transporters were shown to function as obligate hetero-dimers to promote sterol excretion into bile [21]. Overexpression ofABCG5/G8 in transgenic mice results in increased hepatobiliaryexcretion of cholesterol and fecal neutral sterol excretion [21e24],while inactivation of both genes has the opposite effect [25].Although the upregulation of LXR targets ABCG5/8 in EP 80317-treated mice (Fig. 2B) suggested a potential contribution of thesetransporters in cholesterol efflux at the apical site of the enter-ocytes to the lumen, no significant modulation of ABCG8 proteinlevel could be observed (Fig. 3B), so that cautionmust be takenwiththe interpretation of their role in EP 80317-mediated macrophage-to-feces RCT. In contrast, a reduction in NPC1L1 mRNA levels injejunum has been observed following treatment with EP 80317.This reduction of NPC1L1 expression might be LXR driven as pre-viously reported [12], resulting in a substantial reduction in intes-tinal cholesterol absorption and increased fecal cholesterolexcretion [26,27] as observed in NPC1L1-null mice [28], thus pre-venting the development of atherosclerosis in apoE�/� mice[29,30]. As observed in macrophages harvested from EP 80317-treated mice [3], no change was observed in CD36 mRNA levels inthe small intestine following a prolonged (12 weeks) treatmentwith EP 80317 (Fig. 2B). CD36 is co-expressed with NPC1L1 withinthe small intestine and mediates the uptake of both fatty acids andcholesterol mainly in the proximal segment. Interestingly, NPC1L1is upregulated in CD36-deficient tissue, which may account for the

Fig. 3. Intestinal levels of apical intestinal transporters as determined by Western blot. Tissues were obtained from apoE�/� mice treated or not with EP 80317 (300 mg/kg per day)for 6 weeks. Total proteins were extracted from jejunum and were probed for NPC1L1 (A) and ABCG8 (B). Data were normalized to b-actin and presented as fold change (�SEM)relative to vehicle. Numbers of mice jejunums were 7e8. *P < 0.05 compared with 0.9% NaCl-treated mice.

Fig. 4. EP 80317-mediated effects are CD36-dependent. ApoE�/�/CD36�/� mice, pretreated or not with EP 80317 (300 mg/kg per day) for 12 weeks, were injected i.p. with [3H]-cholesterol-loaded J774 macrophages. Plasmaetime curves of [3H]-tracer (0e48 h), expressed as [3H]-tracer % of injection/ml plasma following the injection of radiolabelled-macrophages (A). Area under curve (AUC) is expressed as (pmol � h/ml) (B). Radioactivity content of liver at 48 h (C) and of feces (0e48 h) (D). Values are expressed as [3H]-tracer % of injection/100 mg � SEM (n ¼ 7 and 6 mice in 0.9% NaCl and EP 80317 group, respectively). Small intestines were obtained from apoE�/�/CD36�/� mice treated with orwithout EP 80317. Total RNAwas extracted from small intestines, and mRNA levels of indicated genes were determined by real-time RT-PCR (n ¼ 6e7) (E). In additional studies, totalproteins from small intestines of double knockout mice were assessed for modulation of NPC1L1 and ABCG8 levels by EP 80317 treatment by Western blot (n ¼ 3e4) (F). Data arepresented as fold change (�SEM) versus vehicle.

K. Bujold et al. / Atherosclerosis 229 (2013) 408e414412

K. Bujold et al. / Atherosclerosis 229 (2013) 408e414 413

observation that cholesterol absorption from the intestine iscompensated for the loss of intestinal absorptive activity in CD36deficiency [31]. Further studies are needed to clarify the mecha-nisms of the inhibitory effect of CD36 ligands on NPC1L1 expressionin the intestine.

Among drugs that may modulate RCT, LXR agonists have beenshown to increase RCT and fecal cholesterol excretion by upregu-lating hepatic ABCG5 and ABCG8. Yet, a drawback associated withLXR agonists was the development of liver steatosis [32e34],whereas chronic administration of EP 80317 was not associatedwith fatty liver [3], which is in line with the low [3H]-cholesterolcontent in liver of EP 80317-treated mice. Furthermore, in contrastwith LXR agonists, no change in hepatic mRNA levels of ABCtransporters, CYP7a1 or LXRa was found in the livers of EP 80317-treated mice, and neither LDL nor VLDL plasma clearance wasmodulated following chronic treatment with EP 80317(Supplementary Fig. 1), suggesting that the CD36 ligand does notstimulate hepatic cholesterol metabolism. Thus, EP 80317-mediated decrease in hepatic [3H]-cholesterol might be explainedby reduced cholesterol hepatic uptake as a consequence ofincreased TICE, rather than through a change in lipoprotein frac-tional catabolic rate or hepatic cholesterol transporters such asscavenger receptor class B type I or low density lipoprotein re-ceptor. Alternatively, GW610742, a PPARd agonist, was shown toincrease intestinal cholesterol excretion by decreasing NPC1L1expression rather than upregulating intestinal ABCG5 and ABCG8expression [27,35,36]. In this regard, ezetimibe has been shown toinhibit the function of NPC1L1 in the small intestine and liver [37],fostering the development of NPC1L1 inhibitors as hypocholester-olemic drugs.

To further investigate the molecular mechanisms throughwhich EP 80317 increased TICE, we investigated its effect onintracellular vesicular cholesterol transport proteins Rab8 andRab9. Whereas Rab8 contributes to endosomal cholesterol removal[38], Rab9 plays a role in cholesterol trafficking from late endo-somes to the trans-Golgi network [39,40]. Our results show that EP80317 pretreatment did not alter intestinal Rab8 and Rab9 geneexpression (data not shown). Although the present study did notallow delineation of the molecular mechanisms involved in CD36-dependent increase in TICE following a prolonged treatment withEP 80317, it underscored a potentially novel pharmacological classof drugs which may negatively regulate NPC1L1 expression in aCD36-dependent manner. Further studies will be necessary toelucidate the lipoproteins, cholesterol acceptors or transportersinvolved in the apical and basolateral membrane as well as theroute of cholesterol trafficking in enterocytes.

In conclusion, our results support that a chronic treatment withEP 80317 selectively increases macrophage-to-feces RCT throughthe scavenger receptor CD36 by promoting the alternative RCT, TICEpathway. The mechanism appears to be linked to the LXRa-medi-ated downregulation of intestinal NPC1L1 expression. Thisenhancement of TICE may contribute, at least in part, to thereduction of atherosclerotic lesion formation and hypocholester-olemia following a chronic administration of EP 80317 as a selectiveligand of CD36.

Funding

This work was supported by the Canadian Institutes of HealthResearch (MOP 62837) and by Ardana Bioscience (Edinburgh,Scotland, UK). K. B. was a recipient of a doctoral fellowship fromRobert Dugal-Rx&D Foundation and D. R. was a recipient of apostdoctoral scholarship from the Fonds de la Recherche en Santédu Québec. K.F.Z. is a recipient of a scholarship from the EmergingLeaders in the Americas Program (ELAP), Government of Canada.

Acknowledgements

The authors wish to gratefully acknowledge the skillful technicalassistance of Louise Falstrault, Petra Pohankova and VilayphoneLuangrath.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.atherosclerosis.2013.05.031.

References

[1] Febbraio M, Podrez EA, Smith JD, Hajjar DP, Hazen SL, Hoff HF, et al. Targeteddisruption of the class B scavenger receptor CD36 protects against athero-sclerotic lesion development in mice. J Clin Invest 2000;105:1049e56.

[2] Nagy L, Tontonoz P, Alvarez JG, Chen H, Evans RM. Oxidized LDL regulatesmacrophage gene expression through ligand activation of PPARgamma. Cell1998;93:229e40.

[3] Marleau S, Harb D, Bujold K, Avallone R, Iken K, Wang Y, et al. EP 80317, aligand of the CD36 scavenger receptor, protects apolipoprotein E-deficientmice from developing atherosclerotic lesions. FASEB J 2005;19:1869e71.

[4] HarbD, BujoldK, FebbraioM, SiroisMG,OngH,Marleau S. The role of the scavengerreceptor CD36 in regulating mononuclear phagocyte trafficking to atheroscleroticlesions and vascular inflammation. Cardiovasc Res 2009;83:42e51.

[5] Bujold K, Rhainds D, Jossart C, Febbraio M, Marleau S, Ong H. CD36-mediatedcholesterol efflux is associated with PPARgamma activation via a MAPK-dependent COX-2 pathway in macrophages. Cardiovasc Res 2009;83:457e64.

[6] Glomset JA. The plasma lecithins:cholesterol acyltransferase reaction. J LipidRes 1968;9:155e67.

[7] Cuchel M, Rader DJ. Macrophage reverse cholesterol transport: key to theregression of atherosclerosis? Circulation 2006;113:2548e55.

[8] van der Velde AE, Brufau G, Groen AK. Transintestinal cholesterol efflux. CurrOpin Lipidol 2010;21:167e71.

[9] van derVeldeAE, VrinsCL, van denOeverK, KunneC,OudeElferinkRP,Kuipers F,et al. Direct intestinal cholesterol secretion contributes significantly to total fecalneutral sterol excretion in mice. Gastroenterology 2007;133:967e75.

[10] van der Veen JN, van Dijk TH, Vrins CL, van Meer H, Havinga R, Bijsterveld K,et al. Activation of the liver X receptor stimulates trans-intestinal excretion ofplasma cholesterol. J Biol Chem 2009;284:19211e9.

[11] Abumrad NA, Davidson NO. Role of the gut in lipid homeostasis. Physiol Rev2012;92:1061e85.

[12] Duval C, Touche V, Tailleux A, Fruchart JC, Fievet C, Clavey V, et al. Niemann-Pick C1 like 1 gene expression is down-regulated by LXR activators in theintestine. Biochem Biophys Res Commun 2006;340:1259e63.

[13] Kruit JK, Plosch T, Havinga R, Boverhof R, Groot PH, Groen AK, et al. Increasedfecal neutral sterol loss upon liver X receptor activation is independent ofbiliary sterol secretion in mice. Gastroenterology 2005;128:147e56.

[14] Brown JM, Bell TA, Alger HM, Sawyer JK, Smith TL, Kelley K, et al. Targeteddepletion of hepatic ACAT2-driven cholesterol esterification reveals a non-biliary route for fecal neutral sterol loss. J Biol Chem 2008;283:10522e34.

[15] Temel RE, Brown JM. A new framework for reverse cholesterol transport: non-biliary contributions to reverse cholesterol transport. World J Gastroenterol2010;16:5946e52.

[16] Temel RE, Sawyer JK, Yu L, Lord C, Degirolamo C, McDaniel A, et al. Biliarysterol secretion is not required for macrophage reverse cholesterol transport.Cell Metab 2010;12:96e102.

[17] Vrins CL. From blood to gut: direct secretion of cholesterol via transintestinalcholesterol efflux. World J Gastroenterol 2010;16:5953e7.

[18] Zhang Y, Da Silva JR, Reilly M, Billheimer JT, Rothblat GH, Rader DJ. Hepaticexpression of scavenger receptor class B type I (SR-BI) is a positive regulator ofmacrophage reversecholesterol transport invivo. J Clin Invest2005;115:2870e4.

[19] Werb Z, Chin JR. Onset of apoprotein E secretion during differentiation ofmouse bone marrow-derived mononuclear phagocytes. J Cell Biol 1983;97:1113e8.

[20] Lo SG, Murzilli S, Salvatore L, D’Errico I, Petruzzelli M, Conca P, et al. Intestinalspecific LXR activation stimulates reverse cholesterol transport and protectsfrom atherosclerosis. Cell Metab 2010;12:187e93.

[21] Graf GA, Yu L, Li WP, Gerard R, Tuma PL, Cohen JC, et al. ABCG5 and ABCG8 areobligate heterodimers for protein trafficking and biliary cholesterol excretion.J Biol Chem 2003;278:48275e82.

[22] Yu L, Li-Hawkins J, Hammer RE, Berge KE, Horton JD, Cohen JC, et al. Over-expression of ABCG5 and ABCG8 promotes biliary cholesterol secretion andreduces fractional absorption of dietary cholesterol. J Clin Invest 2002;110:671e80.

[23] Yu L, York J, von Bergman K, Lutjohann D, Cohen JC, Hobbs HH. Stimulation ofcholesterol excretion by the liver X receptor agonist requires ATP-bindingcassette transporters G5 and G8. J Biol Chem 2003;278:15565e70.

[24] Yu L, Gupta S, Xu F, Liverman AD, Moschetta A, Mangelsdorf DJ, et al.Expression of ABCG5 and ABCG8 is required for regulation of biliary choles-terol secretion. J Biol Chem 2005;280:8742e7.

K. Bujold et al. / Atherosclerosis 229 (2013) 408e414414

[25] Yu L, Hammer RE, Li-Hawkins J, von Berkman K, Lutjohann D, Cohen JC, et al.Disruption of Abcg5 and Abcg8 in mice reveals their crucial role in biliarycholesterol secretion. Proc Natl Acad Sci U S A 2002;99:16237e42.

[26] Altmann SW, Davis HR, Zhu LJ, Yao X, Hoos LM, Tetzloff G, et al. Niemann-PickC1 like 1 protein is critical for intestinal cholesterol absorption. Science2004;303:1201e4.

[27] Briand F, Naik SU, Fuki I, Millar JS, Macphee C, Walker M, et al. Both theperoxisome proliferator-activated receptor delta agonist, GW0742, and eze-timibe promote reverse cholesterol transport in mice by reducing intestinalreabsorption of HDL-derived cholesterol. Clin Transl Sci 2009;2:127e33.

[28] Davis HR, Zhu LJ, Hoos LM, Tetzloff G, Maguire M, Liu J, et al. Niemann-Pick C1like 1 (NPC1L1) is the intestinal phytosterol and cholesterol transporter and akey modulator of whole-body cholesterol homeostasis. J Biol Chem 2004;279:33586e92.

[29] Davis HR, Compton DS, Hoos L, Tetzloff G. Ezetimibe, a potent cholesterolabsorption inhibitor, inhibits the development of atherosclerosis in ApoEknockout mice. Arterioscler Thromb Vasc Biol 2001;21:2032e8.

[30] Davis HR, Hoos LM, Tetzloff G, Maguire M, Zhu LJ, Graziano MP, et al. Defi-ciency of Niemann-Pick C1 like 1 prevents atherosclerosis in apoE-/- mice.Arterioscler Thromb Vasc Biol 2007;27:841e9.

[31] Nassir F, Wilson B, Han X, Gross RW, Abumrad NA. CD36 is important for fattyacid and cholesterol uptake by the proximal but not distal intestine. J BiolChem 2007;282:19493e501.

[32] Repa JJ, Liang G, Ou J, Bashmakov Y, Lobaccaro JM, Shimomura I, et al.Regulation of mouse sterol regulatory element-binding protein-1c gene

(SREBP-1c) by oxysterol receptors, LXRalpha and LXRbeta. Genes Dev2000;14:2819e30.

[33] Schultz JR, Tu H, Luk A, Repa JJ, Medina JC, Li L, et al. Role of LXRs in control oflipogenesis. Genes Dev 2000;14:2831e8.

[34] Grefhorst A, Elzinga BM, Voshol PJ, Plosch T, Kok T, Bloks VW, et al. Stimu-lation of lipogenesis by pharmacological activation of the liver X receptorleads to production of large, triglyceride-rich very low density lipoproteinparticles. J Biol Chem 2002;277:34182e90.

[35] Vrins CL, van der Velde AE, van den Oever K, Levels JH, Huet S, Oude Elferink RP,et al. Peroxisome proliferator-activated receptor delta activation leads toincreased transintestinal cholesterol efflux. J Lipid Res 2009;50:2046e54.

[36] van der Veen JN, Kruit JK, Havinga R, Baller JF, Chimini G, Lestavel S, et al.Reduced cholesterol absorption upon PPARdelta activation coincides withdecreased intestinal expression of NPC1L1. J Lipid Res 2005;46:526e34.

[37] Betters JL, Yu L. NPC1L1 and cholesterol transport. FEBS Lett 2010;584:2740e7.[38] Linder MD, Uronen RL, Holtta-Vuori M, van der Sluijs P, Peranen J, Ikonen E.

Rab8-dependent recycling promotes endosomal cholesterol removal innormal and sphingolipidosis cells. Mol Biol Cell 2007;18:47e56.

[39] Narita K, Choudhury A, Dobrenis K, Sharma DK, Holicky EL, Marks DL, et al.Protein transduction of Rab9 in Niemann-Pick C cells reduces cholesterolstorage. FASEB J 2005;19:1558e60.

[40] Choudhury A, Dominguez M, Puri V, Sharma DK, Narita K, Wheatley CL, et al.Rab proteins mediate Golgi transport of caveola-internalized glyco-sphingolipids and correct lipid trafficking in Niemann-Pick C cells. J Clin Invest2002;109:1541e50.