At

JRa

b

c

a

ARRAA

KAISSNOMERa

1

reaumolsr

T

0d

Atherosclerosis 213 (2010) 241–246

Contents lists available at ScienceDirect

Atherosclerosis

journa l homepage: www.e lsev ier .com/ locate /a therosc leros is

nti-inflammatory salicylate beneficially modulates pre-existing atherosclerosishrough quenching of NF-�B activity and lowering of cholesterol

itske de Vries-van der Weija,b, Karin Toeta, Susanne Zadelaara, Peter Y. Wielingaa,obert Kleemanna,∗, Patrick C.N. Rensenc, Teake Kooistraa,∗

The Netherlands Organization for Applied Scientific Research (TNO), Biosciences, Gaubius Laboratory, P.O. Box 2215, 2301 CE Leiden, The NetherlandsThe Dept. of Human Genetics, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The NetherlandsThe Dept. of General Internal Medicine, Endocrinology, and Metabolic Diseases, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The Netherlands

r t i c l e i n f o

rticle history:eceived 6 June 2010eceived in revised form 15 August 2010ccepted 5 September 2010vailable online 21 September 2010

eywords:nti-inflammatory treatment

nflammationtatinerum amyloid A (SAA)F-�Bn-top of statin treatmentouse model of atherosclerosis

stablished lesions

a b s t r a c t

Objective: Inflammation plays an important role in all stages of atherosclerosis, but little is known aboutthe therapeutic effects of quenching inflammation in already existing atherosclerotic lesions. Putativebeneficial effects of salicylate, an inhibitor of NF-�B activation, were studied in mice with establishedlesions.Methods: ApoE*3-Leiden mice received a high-cholesterol diet (HC) to establish atherosclerotic lesions.Reference mice (REF) were sacrificed to determine the lesion area at the start of two interventions. Inone intervention group HC diet feeding was continued, but the diet contained salicylate (HC + SAL). Assalicylate not only quenches inflammation but also reduces plasma cholesterol, a second interventiongroup was fed a low-cholesterol diet (LC) resulting in cholesterol levels comparable to HC + SAL. Theeffects of these interventions on lesion area and composition were assessed after 8 and 16 weeks.Results: HC + SAL markedly reduced hepatic NF-�B activity compared to REF, and was significantly moreeffective than LC diet feeding. HC + SAL and LC also quenched aortic NF-�B activity. While continuing HCdiet typically further increases total lesion area, 16 weeks of intervention with HC + SAL and LC halted

educe further progression oftherosclerosis

further disease progression and resulted in lesion sizes comparable to that of REF. At the same time, lesioncomposition was significantly improved, particularly with salicylate. Strikingly, HC + SAL resulted in alower lesional macrophage content and a greater plaque stability index (ratio of collagen to macrophagearea) than LC.Conclusion: Anti-inflammatory salicylate reduces atherosclerotic macrophage content and increaseslesion stability of pre-existing plaques through quenching of NF-�B activity and reducing plasma choles-

terol.

. Introduction

Hypercholesterolemia is widely recognized as an importantisk factor for atherosclerosis development [1]. It is also well-stablished that inflammation plays a role in all stages of thetherosclerotic process, from lesion initiation to progression andltimately to plaque rupture and thrombosis [2,3]. Numerous ani-al studies have shown that targeting either hypercholesterolemia

r inflammation can prevent the formation of new atheroscleroticesions [4–7]. However, different from these animal studies, humanubjects eligible for treatment already have established atheroscle-otic lesions.

∗ Corresponding authors. Tel.: +31 71 5181450; fax: +31 71 5181901.E-mail addresses: Robert.Kleemann@tno.nl (R. Kleemann),

eake.Kooistra@tno.nl (T. Kooistra).

021-9150/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.atherosclerosis.2010.09.006

© 2010 Elsevier Ireland Ltd. All rights reserved.

The development of treatment strategies that promote remod-eling and eventually regression of pre-existing lesions is thereforean important issue, but has been insufficiently addressed. Animalstudies have shown that reducing hypercholesterolemia stronglydecreases the number of lesional macrophage foam cells [8–10]within a few days [11]. Later in time, a reduction of plaque sizeis observed [8,9,11]. In humans, non-invasive imaging confirmedthat plasma cholesterol-lowering alone can promote atheroscle-rosis regression [12]. Recent findings also suggest that targetingthe inflammatory component of atherosclerotic disease may have abeneficial effect on pre-existing lesions [10,13]. However, an inter-vention with an anti-inflammatory drug has so far not been tested

in a setting with pre-existing lesions.

Salicylate (SAL) is an anti-inflammatory drug that inhibits theactivity of nuclear factor-�B (NF-�B), a key mediator of the inflam-matory response. SAL acts by quenching the activity of the inhibitorof �B (I�B) kinase (IKK) complex [14], thereby preventing the

2 Athero

ptat

aaeiivbmseipctlea

2

2

aaWtimhr(1rl(mLa

2l

(fiÄca

2

ttRsMaod

42 J. de Vries-van der Weij et al. /

hosphorylation of I�B and the subsequent release and nuclearranslocation of NF-�B. Also, in addition to its anti-inflammatoryction, we have recently reported that SAL can lower plasma choles-erol levels in mice [15].

In the present study we tested whether quenching of NF-�Bctivity with SAL can contribute to remodeling of pre-existingtherosclerotic lesions on top of its cholesterol-lowering effect. Toxamine this, we used APOE*3-Leiden (E3L) mice, which respondn a human-like manner to standard anti-atherosclerotic drugs andn which plasma cholesterol can be titrated to a desired stable levelia the diet [10,16]. After establishment of atherosclerotic lesionsy feeding E3L mice a high-cholesterol (HC) diet for 19 weeks, theice were either sacrificed (reference group; REF) or fed HC diet

upplemented with SAL for another 8 or 16 weeks to assess theffect of SAL on pre-existing plaques. Notably, continuing feed-ng the mice HC diet alone has previously been shown to furtherromote lesion progression [7]. As SAL treatment lowers plasmaholesterol, a control group received a diet with a lower choles-erol content (LC diet) to achieve comparable plasma cholesterolevels as with SAL. This enabled us to study the anti-inflammatoryffects of SAL independent of cholesterol-lowering on pre-existingtherosclerotic lesions over time.

. Methods

.1. Animals

12- to 14-week-old female E3L mice were housed with freeccess to food and water (12-h light–dark cycle). Mice were fedHC diet containing 15% (w/w) cacao butter (diet T, Hope Farms,oerden, The Netherlands) supplemented with 1.0% (w/w) choles-

erol (Sigma) for 19 weeks. After this period, mice were randomizednto three groups and matched for plasma cholesterol. One group of

ice (n = 15) was sacrificed in week 19 (CO2 inhalation) to collectearts and livers (reference group; REF). A second group (n = 30)eceived the HC diet supplemented with 0.4% (w/w) salicylic acidcat. no. 24,758-8, Aldrich, Steinheim, Germany) for 8 weeks or6 weeks (n = 15 sacrificed per time point). A third group (n = 30)eceived an LC diet (with composition exactly as HC but containingess (0.12% w/w) cholesterol), also for 8 weeks (n = 15) or 16 weeksn = 15). Fasting 4-h blood was collected monthly. All animal experi-

ents were performed according to the Guide for the Care and Use ofaboratory Animals (NIH-Publication No. 85-23, revised 1996) andpproved by the institutional Animal Experimental Committee.

.2. Analysis of plasma cholesterol, cholesterol distribution overipoproteins and inflammatory markers

Plasma total cholesterol was measured using kit no. 1489437Roche Diagnostics, Mannheim, Germany). For lipoprotein pro-les, plasma was pooled per group and fractionated using anKTA fast protein liquid chromatography (FPLC) system (Pharma-ia, Roosendaal, The Netherlands) [17]. Plasma levels of serummyloid A (SAA) were determined by ELISA (Tridelta, Ireland) [18].

.3. Analysis of NF-�B and STAT 3 activities in liver and aorta

Relative amounts of p65-NF-�B and STAT3 transcription fac-or activity in livers and aortas were determined using TransAMranscription factor assay kit nos. 40097 and 45196 (Active Motif,ixensart, Belgium) as reported [28]. Homogenates of each tis-

ue were prepared with Nuclear Extract Kit (no. 40010, Activeotif) and samples were taken to determine protein content. Equal

mounts of protein were used in the TransAM assays. Bindingf the active transcription factor to a consensus sequence wasetermined in the presence of either a competitive or a mutated

sclerosis 213 (2010) 241–246

(non-competitive) oligonucleotide to correct for non-specific bind-ing.

2.4. Atherosclerosis quantification

Atherosclerotic lesion area as well as lesional macrophage andcollagen content were determined as reported [19]. Briefly, heartswere fixed in phosphate-buffered 4% formaldehyde, dehydrated,embedded in paraffin, and cross-sectioned (aortic root area). Foreach mouse, four sections with 50 �m intervals were stained withhematoxylin–phloxin–saffron (HPS) for histological analysis usingCell D imaging software (Olympus Soft Imaging Solutions). AIA31240 antiserum (1:3000, Accurate Chemical and Scientific) andSirius Red were used to quantify macrophage area and collagenarea, respectively [19]. The plaque stability index was determinedby dividing the collagen area by the macrophage area.

2.5. Statistical analysis

Significance of difference was tested by one-way analysis ofvariance (ANOVA) test followed by a least significant differencepost hoc analysis using SPSS 16.0 for Windows (SPSS, Chicago, USA).

3. Results

3.1. Salicylate treatment and LC diet feeding reduce plasmaVLDL-cholesterol comparably

E3L mice were fed a high-cholesterol (HC) diet to establishatherosclerotic lesions. HC diet feeding increased plasma choles-terol levels from approximately 2 mM to 15.4 ± 2.7 mM (Fig. 1A).After 19 weeks, the mice were randomized into three groups. Onegroup was sacrificed immediately and served as a reference (REF).Another group received HC diet supplemented with 0.4% (w/w)salicylate (HC + SAL group), which lowered plasma cholesterol lev-els to 7.9 ± 1.5 mM. The third group received a low-cholesteroldiet (LC group), which resulted in a comparable plasma choles-terol of 7.6 ± 1.7 mM. The plasma cholesterol levels of HC + SAL andLC remained comparable at all time points analyzed. Lipoproteinprofiling showed that the cholesterol distribution was similar inHC + SAL and LC groups and that the observed decrease in plasmacholesterol was related to a comparable relative reduction of VLDLand LDL (Fig. 1B). After prolonged (8 or 16 weeks) HC + SAL or LCdiet feeding the mice were sacrificed and livers, hearts and thoracicaortas were isolated for further analysis.

3.2. Salicylate quenches hepatic NF-�B activity beyond the effectof LC diet feeding

To evaluate the effect of salicylate on hepatic inflammation,plasma levels of the hepatic inflammation marker SAA were mea-sured. HC feeding increased SAA from 5.4 ± 0.4 �g/mL at baseline(t = 0) to 10.3 ± 3.9 �g/mL in week 19 (Fig. 2A). Subsequent inter-vention with HC + SAL reduced SAA to 8.0 ± 3.9 �g/mL within 8weeks (P < 0.05) and to 7.6 ± 3.3 �g/mL after 16 weeks (P < 0.01). LCdiet reduced SAA levels also, to 6.2 ± 2.0 �g/mL in week 8 (P < 0.001)and 5.5 ± 2.0 �g/mL in week 16 (P < 0.001). The SAA reduction bycholesterol-lowering alone (LC group) was greater than with sali-cylate, an effect that became significant in week 16 (P < 0.05).

To verify that salicylate treatment reduced hepatic NF-�B activ-ity, livers were homogenized and assayed using TransAM assays.

NF-�B activity in REF mice was set 100 (Fig. 2B). Compared toREF, treatment with HC + SAL strongly quenched NF-�B activity(−75% in week 8, P < 0.001), which remained low (−63% in week 16,P < 0.001). LC diet reduced NF-�B activity less strong than salicylate(significant difference between groups P < 0.01), both after 8 weeks

J. de Vries-van der Weij et al. / Atherosclerosis 213 (2010) 241–246 243

Sacrifice reference HC group (REF),

start interventionA

15

20HC(REF)

LC

HC+SAL

***

5

10 ***

******

***

******

******

***

0 10 20 30 40

0

Time (weeks)

Pla

sm

a c

ho

leste

rol (m

M)

B

2.0 HDLLDL/HDL1VLDL

1.0

1.5

0.5

Ch

ole

ste

rol (m

M)

0 5 10 15 20 25

0.0

Fraction (#)

Fig. 1. Effect of salicylate and low-cholesterol diet feeding on plasma lipids. E3L micewere fed a high-cholesterol (HC) diet for 19 weeks and then randomized into threegroups. One group was immediately sacrificed and served as a reference (REF). Theother groups received either HC diet but supplemented with salicylate (HC + SAL) or alow-cholesterol (LC) diet for an additional 8 or 16 weeks. (A) Plasma total cholesterolover time. Values are means ± SD; ***P < 0.001 compared to REF. (B) Pooled plasmasclt

(tddttbSb

3r

tHiow

ttf

10

15

HC

HC+SAL

***** *

### $A

REF

0

5

LC

### ###

Pla

sma

SA

A

(μg

/ml)

****

**

0 10 20 30 40

Time (weeks)

B

80

100

120

140 ***

** **

LCHC+SALLCHC+SALHC0

20

40

60

Hep

atic

NF

-κB

act

ivit

y (A

U)

+ 16 weeks19 weeks + 8 weeks

160*

**C

60

80

100

120

140

LCHC+SALLCHC+SALHC0

20

40

+ 16 weeks19 weeks + 8 weeks

Hep

atic

ST

AT

3 a

ctiv

ity

(A

U)

Fig. 2. Effect of salicylate and LC diet feeding on hepatic inflammation markersand intrahepatic NF-�B activity. (A) Plasma samples were analyzed for SAA at the

#

amples were fractionated by FPLC and the individual fractions were assayed forholesterol at different time points during the course of the study. Representativeipoprotein profiles of one time point are shown (REF: t = week 19; HC + SAL and LC:= week 27).

−33%, P < 0.01) and 16 weeks (−27%, P < 0.05). To assess whetherhe more pronounced quenching of NF-�B activity in HC + SAL wasue to a general anti-inflammatory effect of salicylate, we alsoetermined the activity of STAT3, another key transcription factorhat controls the expression of inflammatory genes [20]. Comparedo the REF, cholesterol-lowering alone in LC reduced STAT3 activityy 42% (P < 0.05; 8 weeks) and by 39% (P < 0.05; 16 weeks) (Fig. 2C).alicylate had a less pronounced effect and reduced STAT3 activityy 36% in week 8 (P < 0.05) and by 16% in week 16 (n.s.).

.3. Salicylate beneficially affects pre-existing atherosclerosis byeducing macrophage area and by increasing lesion stability

To evaluate if salicylate also quenches NF-�B activity in aorta,horacic aorta homogenates were prepared. Compared to REF,C + SAL showed a significantly lower aortic NF-�B activity (−51%

n week 8 and −53% in week 16; both P < 0.01) (Fig. 3A). The effectf cholesterol-lowering alone in LC was less pronounced (−32% in

eek 8 and −34% in week 16, P < 0.05).

Analysis of atherosclerotic lesions in the aortic root showedhat REF developed mainly mild atherosclerotic lesions with aotal lesion size of 52 ± 31 × 103 �m2. While continuing HC dieteeding has been reported to result in development of large and

time points indicated; *P < 0.05, **P < 0.01, ***P < 0.001 compared to t = 0; P < 0.05,##P < 0.01, ###P < 0.001 compared to HC in week 19 (REF); $P < 0.05 compared toLC. (B) Liver homogenates were assayed for active p65-NF-�B and (C) STAT3 usingTransAM assays. Values are means ± SD; *P < 0.05, **P < 0.01, ***P < 0.001.

advanced lesions [7], total lesion area did not significantly furtherincrease over time in both the LC group and the HC + SAL group.With LC, the total lesion area was 77 ± 66 × 103 �m2 (week 8) and60 ± 50 × 103 �m2 (week 16) and thus slightly higher than at thestart of the intervention in REF. Salicylate treatment, by contrast,resulted in a total lesion area of 65 ± 49 × 103 �m2 (week 8) and42 ± 41 × 103 �m2 (week 16), a slightly lower area than in REF(Fig. 3C).

Next, the effect of salicylate on the lesional macrophage area andcollagen content was investigated (see Fig. 3B for representativemorphological changes, and Fig. 3D and E for planimetry data). LCfeeding reduced the macrophage area with 69% compared to theREF group (P < 0.001), and this reduction persisted (−61% in week16; P < 0.001). Salicylate treatment reduced lesional macrophagearea even more than LC (−87% in week 8 and −92% in week 16; bothP < 0.001). The difference in macrophage area between HC + SAL andLC was significant in week 16 (P < 0.05).

The relative lesional collagen content increased in the LC andHC + SAL groups compared to the REF group. The plaque stabil-ity index, calculated as the ratio of collagen to macrophage area,strongly increased over time in LC (P < 0.01) and HC + SAL (P < 0.001)

244 J. de Vries-van der Weij et al. / Atherosclerosis 213 (2010) 241–246

Fig. 3. Salicylate reduces lesional macrophage area and increases lesion stability more than cholesterol-lowering alone. (A) Aortas were homogenized and active p65-NF-�B was determined by TransAM analysis. (B) Hearts were embedded in paraffin and cross-sectioned throughout the aortic root area. Sections were stained withhematoxylin–phloxin–saffron (HPS; histological analysis), AIA 31240 antiserum (for macrophages) and Sirius Red (for collagen). Representative photomicrographs of theaortic root area of REF, HC + SAL and LC are shown. (C; see next page) Total atherosclerotic lesion areas of the experimental groups (points represent individual mice). Hori-zontal bars indicate mean lesion area. (D) Lesional macrophage content. (E) Plaque stability index (ratio of collagen to macrophage area of the lesions). Values are means ± SD;*P < 0.05, **P < 0.01, ***P < 0.001.

J. de Vries-van der Weij et al. / Atherosclerosis 213 (2010) 241–246 245

(Conti

cmL

4

tra[pilEacirTisc

csalOadoisi

Fig. 3.

ompared with REF. Importantly, the plaque stability index wasarkedly higher in salicylate treated mice compared to mice fed

C in week 8 and 16 (P < 0.05).

. Discussion

It is well recognized that hypercholesterolemia and inflamma-ion are important risk factors in the development of atheroscle-osis. While it has been shown that plasma cholesterol-loweringlone can induce remodeling of pre-existing atherosclerotic lesions8,9,11], the effect of quenching inflammation on pre-existinglaques has received as yet little attention. In this study, we

nvestigated the effect of quenching NF-�B activity with salicy-ate on established atherosclerotic lesions in high cholesterol fed3L mice. Because salicylate, in addition to reducing hepatic andortic NF-�B activity, also lowers plasma cholesterol levels, aholesterol-matched group was included (LC). An important find-ng is that quenching of NF-�B activity with salicylate stronglyeduced macrophage area and increased plaque stability over time.hese effects were more pronounced in the HC + SAL group thann the LC group, indicating that the greater benefit achieved withalicylate is not simply a consequence of its cholesterol-loweringapacity.

A decrease in lesional macrophage content following aholesterol-lowering regimen as demonstrated in the presenttudy has also been observed by others. Trogan et al. [11] reportedstrong decrease in the number of macrophages from existing

esions already after a few days of lowering plasma cholesterol.ur finding that salicylate accelerates macrophage disappear-nce beyond the effect of cholesterol-lowering alone is new and

eserves further attention. Concomitant with the disappearancef macrophages, we show that the plaque stability index stronglyncreased during the intervention with LC and even more so withalicylate. This is in line with studies that observed an increasen the collagen-producing cell types, fibroblasts [8] or smooth

nued ).

muscle cells (SMCs), under atherosclerosis regression conditions[9,21]. Notably, and similar to the data on the disappearance ofmacrophages, the effects of salicylate on the stability of pre-existingatherosclerotic lesions are stronger than those of cholesterol-lowering per se.

We investigated the effect of quenching NF-�B activity onestablished atherosclerotic lesions under conditions of low plasmacholesterol. While, to our knowledge, there are no other reportedstudies evaluating the role of NF-�B in remodeling of pre-existingatherosclerosis, several studies addressed the role of NF-�B in thedevelopment of new lesions. However, results so far have notbeen unequivocal. Studies investigating the effect of macrophage-specific reduction of NF-�B activity on atherosclerosis developmentin hyperlipidemic LDLr−/− mice showed increased numbers oflesional inflammatory cells [22,23]. This finding was thought to bethe consequence of reduced lipid uptake by macrophages, leadingto an inflammatory response of the non-phagocytosed lipids andconsequently to enhanced recruitment of leukocytes [23]. How-ever, in another study, using a genetic apolipoprotein E-deficient(ApoE−/−) mouse model, diminished atherosclerosis and decreasednumbers of inflammatory cells in the lesions were found whendecreasing NF-�B activity [24]. Also, a study on the effect of acetyl-salicylic acid (aspirin) on atherogenesis in LDLr−/− mice showedthat aspirin reduced NF-�B activity in aortic tissue. This was asso-ciated with reduced lesion size and lesional macrophages, but anincrease in lesional smooth muscle cells and collagen, thus reflect-ing increased plaque stability [25]. Since aspirin also quenchesinflammation by reducing prostaglandin synthesis, the specificcontribution of NF-�B inhibition to the observed effects remaineduncertain. The latter two studies showed a beneficial effect of

quenching NF-�B activity under high plasma cholesterol condi-tions. Notably, in these studies, whole body NF-�B activity wasquenched while in the studies of Kanters et al. [22,23] showingincreased numbers of inflammatory cells in atherosclerotic lesions,macrophage-specific reduction of NF-�B activity was evaluated.

2 Athero

Cmt

ttihcosetwSgosalp

oiiittmp

isccmbdvicp

F

CbtRF

C

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

expression. Blood 2004;103:413–21.

46 J. de Vries-van der Weij et al. /

ollectively, these data suggest that quenching of overall, but notacrophage-specific NF-�B activity can have beneficial effects on

he composition of both newly developing and pre-existing lesions.The mechanism by which quenching of NF-�B activity reduces

he lesional macrophage content is as yet unknown. We reasonhat this mechanism may involve increased turnover rather thanncreased emigration of macrophages. Quenching NF-�B activityas been reported to coincide with decreased expression of C–Chemokine receptor 7 (CCR7) [26], which is required for migrationf dendritic cell-like macrophages out of existing lesions [11]. Con-equently, quenching NF-�B activity would be expected to decreasemigration of monocytes from lesions. On the other hand, inhibi-ion of NF-�B activity has been reported to induce apoptosis, whichould be in line with diminished lesional macrophage content [27].

upport for this notion also comes from a study evaluating athero-enesis in mice lacking IKK2 specifically in macrophages. Deletionf IKK2 diminished NF-�B activation, and atherosclerotic lesionshowed increased cell death [22]. Therefore, increased apoptosiss a result of diminished NF-�B activity may underlie the loweresional macrophage content of salicylate-treated mice as com-ared to LC fed mice.

Treatment with LC and HC + SAL diets had a comparable effectn the total lesion area, yet marked differences were found regard-ng plaque composition and hepatic NF-�B activation which weremproved by HC + SAL treatment. Hepatic NF-�B activity was signif-cantly reduced with HC + SAL when compared to LC treatment buthere was no statistical difference in aortic NF-�B activity betweenhese groups. Therefore, the present study does not allow esti-

ation of the effect of reducing aortic NF-�B inflammation byharmacological means separate from that of diet.

The clinical importance of gaining insight in processes involvedn remodeling of pre-existing atherosclerosis that lead to plaquetabilization or regression is increasingly recognized. Our findingsontribute to this insight. We show that salicylate-induced plasmaholesterol-lowering and quenching of NF-�B activity reducesacrophage content and increases stability of pre-existing lesions

eyond the effects seen with cholesterol-lowering alone. Theseata also illustrate that targeting multiple factors has additionalalue for treatment of (pre-existing) atherosclerosis. Future stud-es are needed to validate whether a combined reduction of plasmaholesterol and NF-�B activity is also beneficial for remodeling ofre-existing lesions in humans.

unding

This work was performed within the framework of the Leidenenter for Cardiovascular Research LUMC-TNO and was supportedy the Centre for Medical Systems Biology [Grant No. 115] withinhe framework of the Netherlands Genomics Initiative (NGI). P.C.N.ensen is an Established Investigator of the Netherlands Heartoundation (Grant NHS 2009T038).

onflict of interest

None.

eferences

[1] Steinberg D. Hypercholesterolemia and inflammation in atherogenesis: twosides of the same coin. Mol Nutr Food Res 2005;49:995–8.

[

[

sclerosis 213 (2010) 241–246

[2] Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation2002;105:1135–43.

[3] Lusis AJ. Atherosclerosis. Nature 2000;407:233–41.[4] Davis Jr HR, Compton DS, Hoos L, et al. Ezetimibe, a potent cholesterol absorp-

tion inhibitor, inhibits the development of atherosclerosis in ApoE knockoutmice. Arterioscler Thromb Vasc Biol 2001;21:2032–8.

[5] Kleemann R, Princen HM, Emeis JJ, et al. Rosuvastatin reduces atherosclero-sis development beyond and independent of its plasma cholesterol-loweringeffect in APOE*3-Leiden transgenic mice: evidence for antiinflammatory effectsof rosuvastatin. Circulation 2003;108:1368–74.

[6] Mach F, Schonbeck U, Sukhova GK, et al. Reduction of atherosclerosis in miceby inhibition of CD40 signalling. Nature 1998;394:200–3.

[7] Verschuren L, Kleemann R, Offerman EH, et al. Effect of low dose atorvastatinversus diet-induced cholesterol lowering on atherosclerotic lesion progressionand inflammation in apolipoprotein E*3-Leiden transgenic mice. ArteriosclerThromb Vasc Biol 2005;25:161–7.

[8] Gijbels MJ, van der Cammen M, van der Laan LJ, et al. Progression and regressionof atherosclerosis in APOE3-Leiden transgenic mice: an immunohistochemicalstudy. Atherosclerosis 1999;143:15–25.

[9] Tsukamoto K, Tangirala R, Chun SH, et al. Rapid regression of atherosclero-sis induced by liver-directed gene transfer of ApoE in ApoE-deficient mice.Arterioscler Thromb Vasc Biol 1999;19:2162–70.

10] Verschuren L, de Vries-van der Weij, Zadelaar S, et al. LXR agonist suppressesatherosclerotic lesion growth and promotes lesion regression in apoE*3Leidenmice:time course and mechanisms. J Lipid Res 2009;50:301–11.

11] Trogan E, Feig JE, Dogan S, et al. Gene expression changes in foam cells and therole of chemokine receptor CCR7 during atherosclerosis regression in ApoE-deficient mice. Proc Natl Acad Sci USA 2006;103:3781–6.

12] Gottlieb I, Agarwal S, Gautam S, et al. Aortic plaque regression as determinedby magnetic resonance imaging with high-dose and low-dose statin therapy. JCardiovasc Med (Hagerstown) 2008;9:700–6.

13] Bernhagen J, Krohn R, Lue H, et al. MIF is a noncognate ligand of CXCchemokine receptors in inflammatory and atherogenic cell recruitment. NatMed 2007;13:587–96.

14] Yin MJ, Yamamoto Y, Gaynor RB. The anti-inflammatory agents aspirin andsalicylate inhibit the activity of I(kappa)B kinase-beta. Nature 1998;396:77–80.

15] Kleemann R, Verschuren L, van Erk MJ, et al. Atherosclerosis and liverinflammation induced by increased dietary cholesterol intake: a combinedtranscriptomics and metabolomics analysis. Genome Biol 2007;8:R200.

16] Zadelaar S, Kleemann R, Verschuren L, et al. Mouse models for atheroscle-rosis and pharmaceutical modifiers. Arterioscler Thromb Vasc Biol2007;27:1706–21.

17] Westerterp M, de Haan W, Berbee JF, Havekes LM, Rensen PC. EndogenousapoC-I increases hyperlipidemia in apoE-knockout mice by stimulating VLDLproduction and inhibiting LPL. J Lipid Res 2006;47:1203–11.

18] Kooistra T, Verschuren L, de Vries-van der Weij, et al. Fenofibrate reducesatherogenesis in ApoE*3Leiden mice: evidence for multiple antiatherogeniceffects besides lowering plasma cholesterol. Arterioscler Thromb Vasc Biol2006;26:2322–30.

19] de Vries-van der Weij J, Zadelaar S, Toet K, et al. Human CETP aggravatesatherosclerosis by increasing VLDL-cholesterol rather than by decreasing HDL-cholesterol in APOE*3-Leiden mice. Atherosclerosis 2009;206:153–8.

20] Pfitzner E, Kliem S, Baus D, et al. The role of STATs in inflammation and inflam-matory diseases. Curr Pharm Des 2004;10:2839–50.

21] Viles-Gonzalez JF, Fuster V, Corti R, et al. Atherosclerosis regression and TPreceptor inhibition: effect of S18886 on plaque size and composition—a mag-netic resonance imaging study. Eur Heart J 2005;26:1557–61.

22] Kanters E, Pasparakis M, Gijbels MJ, et al. Inhibition of NF-kappaB activationin macrophages increases atherosclerosis in LDL receptor-deficient mice. J ClinInvest 2003;112:1176–85.

23] Kanters E, Gijbels MJ, van der Made I, et al. Hematopoietic NF-kappaB1 defi-ciency results in small atherosclerotic lesions with an inflammatory phenotype.Blood 2004;103:934–40.

24] Wolfrum S, Teupser D, Tan M, et al. The protective effect of A20 on atheroscle-rosis in apolipoprotein E-deficient mice is associated with reduced expressionof NF-kappaB target genes. Proc Natl Acad Sci USA 2007;104:18601–6.

25] Cyrus T, Sung S, Zhao L, et al. Effect of low-dose aspirin on vascularinflammation, plaque stability, and atherogenesis in low-density lipoproteinreceptor-deficient mice. Circulation 2002;106:1282–7.

26] Chen T, Guo J, Yang M, et al. Cyclosporin A impairs dendritic cell migrationby regulating chemokine receptor expression and inhibiting cyclooxygenase-2

27] Bellas RE, FitzGerald MJ, Fausto N, et al. Inhibition of NF-kappa B activity inducesapoptosis in murine hepatocytes. Am J Pathol 1997;151:891–6.

28] Kleemann R, van Erk M, Verschuren L, et al. Time-resolved and tissue-specific systems analysis of the pathogenesis of insulin resistance. PLoS One2010;5(1):e8817.