Biochemical Pharmacology 84 (2012) 1627–1634

The PPARd-mediated inhibition of angiotensin II-induced premature senescencein human endothelial cells is SIRT1-dependent

Min Young Kim a,1, Eun Sil Kang a,1, Sun Ah Ham b,1, Jung Seok Hwang b, Tae Sik Yoo b, Hanna Lee b,Kyung Shin Paek c, Chankyu Park b, Hoon Taek Lee b, Jin-Hoi Kim b, Chang Woo Han d, Han Geuk Seo b,*a Department of Pharmacology, Gyeongsang National University School of Medicine, Jinju 660-701, Republic of Koreab Department of Animal Biotechnology, Konkuk University, Hwayang-Dong, Gwangjin-Gu, Seoul 143-701, Republic of Koreac Department of Nursing, Semyung University, Sinwoul-Dong, Jechon, Chungbuk 390-711, Republic of Koread Department of Internal Medicine, Pusan National University School of Korean Medicine, Yangsan, Kyeongnam 626-870, Republic of Korea

A R T I C L E I N F O

Article history:

Received 4 June 2012

Accepted 13 September 2012

Available online 20 September 2012

Keywords:

Angiotensin II

Peroxisome proliferator-activated receptor

d

Premature senescence

Reactive oxygen species

SIRT1

A B S T R A C T

Cellular senescence has been implicated in endothelial dysfunctions affecting vascular tone and

regeneration. The molecular mechanisms of vascular senescence are poorly understood. The present

study demonstrates that upregulation of SIRT1 by peroxisome proliferator-activated receptor (PPAR) dattenuates premature senescence in angiotensin (Ang) II-treated human coronary artery endothelial

cells (HCAECs). Activation of PPARd by the specific ligand GW501516 significantly inhibited Ang II-

induced premature senescence and generation of reactive oxygen species (ROS) in HCAECs. A marked

concentration- and time-dependent increase in the mRNA levels of SIRT1 was observed in GW501516-

treated HCAECs. The effects of GW501516 were almost completely abolished in the presence of small

interfering (si) RNA against PPARd, indicating that PPARd mediates the effects of GW501516. In addition,

activation of PPARd, but not PPARa or PPARg, significantly enhanced SIRT1 promoter activity and

protein expression. Down-regulation or inhibition of SIRT1 by siRNA or sirtinol abrogated the effects of

PPARd on Ang II-induced premature senescence and ROS generation, respectively. Furthermore,

resveratrol, a well-known activator of SIRT1, mimicked the action of PPARd on Ang II-induced premature

senescence and ROS generation. Taken together, these results indicate that the anti-senescent activities

of PPARd may be achieved at least in part by fine tuning the expression of SIRT1 in the vascular

endothelium.

� 2012 Elsevier Inc. All rights reserved.

Contents lists available at SciVerse ScienceDirect

Biochemical Pharmacology

jo u rn al h om epag e: ww w.els evier .c o m/lo cat e/b io c hem p har m

1. Introduction

Vascular endothelium plays a pivotal role in the control ofvascular homeostasis [1]. In physiological environments, vascularendothelial cells are quiescent, divide rarely and have a longturnover rate. In contrast, endothelial cell division is enhanced inconditions that cause endothelium injury, such as hypertensionand high cholesterol, to regenerate the damaged endothelium[2,3]; however the regenerated endothelial cells undergo replica-tive senescence following a finite number of cell division [4].Senescent endothelial cells cause endothelial dysfunction and

Abbreviations: Ang II, angiotensin II; HCAECs, human coronary artery endothelial

cells; PPAR, peroxisome proliferator-activated receptor; PPRE, PPAR response

elements; ROS, reactive oxygen species; SA-b-gal, senescence-associated b-

galactosidase; SIPS, stress-induced premature senescence; sir2, silent information

regulator 2; TGF, transforming growth factor; VSMCs, vascular smooth muscle cells.

* Corresponding author. Tel.: +82 2450 0428; fax: +82 2455 1044.

E-mail address: hgseo@konkuk.ac.kr (H.G. Seo).1 These authors contributed equally to this work.

0006-2952/$ – see front matter � 2012 Elsevier Inc. All rights reserved.

http://dx.doi.org/10.1016/j.bcp.2012.09.008

eventually contribute to the development of age-related cardio-vascular disorders, such as atherosclerosis [1,3]. In fact, vascularcells exhibiting features of cellular senescence have been detectedin human atherosclerosis lesions [5]. In addition, some stressors,including chemical agents [6] and oxidative stress [7], elicit asimilar growth arrest, termed stress-induced premature senes-cence (SIPS), in primary cultured cells. Oxidative stress, caused bythe generation of reactive oxygen species (ROS), has beenimplicated in many senescence-associated vascular diseases [4].ROS break DNA strands and modify bases to elicit both replicativesenescence and SIPS [8]. In fact, increased levels of ROS aredetected in atherosclerotic plaques [5], suggesting a role for ROS incellular senescence-associated vascular disorders [4]. Accordingly,ROS may be responsible for senescence and age-related diseases.

Peroxisome proliferator-activated receptor (PPAR) d is a ligand-activated transcription factor involved in multiple biologicalfunctions [9,10]. This receptor regulates gene expression byforming dimers with the retinoid X receptor via PPAR responseelements (PPRE) located in the regulatory regions of target genes[9]. Recently, it was postulated that ligand-activated PPARd exerts

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–16341628

antiatherosclerotic effects via anti-inflammatory or anti-senescentmechanisms in vascular cells [10–12]. Activation of PPARdsuppresses the inflammation and proliferation of vascular smoothmuscle cells (VSMCs) through the upregulation of transforminggrowth factor (TGF)-b1, an anti-inflammatory mediator [13].Furthermore, we have recently demonstrated that ligand-activatedPPARd counteracts the generation of ROS in VSMCs to inhibit thecellular senescence induced by angiotensin (Ang) II [14]. Based onits beneficial properties in the regulation of vascular inflammationand senescence [11–16], it is particularly important to elucidatewhether PPARd exerts anti-aging effects in endothelial cells.

SIRT1, a mammalian orthologue of yeast sir2 (silent informa-tion regulator 2), is a NAD+-dependent class III proteindeacetylase [17]. SIRT1 has emerged as a vital regulator ofcellular functions including aging, cell cycle, metabolism andapoptosis by interacting with a wide range of substrates such asp53, forkhead transcription factor, MyoD, p53, PGC-1a andhistones [18–21]. In mammalian systems, beneficial effects ofSIRT1 in age-related vascular disorders have been reported in anumber of recent studies [22–25]. In addition, a recent reportshowed that activation of PPARd by a specific ligand increasesSIRT1 promoter activity via sp1 in human hepatocyte-derivedcells [26]. Therefore, we hypothesized that the protective effectsof PPARd against endothelial senescence may be in partattributed to the upregulation of SIRT1. The present resultsstrongly suggest that increased expression of SIRT1 by PPARdmediates the activity of this nuclear factor in the inhibition ofendothelial senescence.

2. Materials and methods

2.1. Materials

Angiotensin II, GW501516, WY14643, troglitazone, resveratroland 20,70-dichlorofluorescein diacetate (H2DCF-DA) wereobtained from Calbiochem (La Jolla, CA, USA). Sirtinol, senescentcell staining kit, and anti-b-actin antibody were purchased fromSigma–Aldrich Co. (St. Louis, MO, USA). The luciferase assaysystem and the b-galactosidase enzyme assay kit were obtainedfrom Promega (Madison, WI, USA). 5-[[4-(2-[Methyl-2-pyridiny-lamino]ethoxy)phenyl]methyl]-2,4-thiazolidinedione (rosiglita-zone) was obtained from Cayman Chemical Company (Ann Arbor,MI, USA). Polyclonal antibodies specific for PPARd and Sirt-1, aswell as horseradish peroxidase (HRP)-conjugated IgG, werepurchased from Santa Cruz Biotechnology (Santa Cruz, CA,USA). Other reagents were of the highest grade available.

2.2. Cell culture

Human coronary artery endothelial cells (HCAECs) werepurchased from ScienCell Research Laboratories (Carlsbad, CA,USA) and cultured in HCAEC growth medium containing endothe-lial cell growth supplements, based on the manufacturer’srecommendations, at 37 8C in an atmosphere of 95% air and 5% CO2.

2.3. Senescence-associated b-galactosidase (SA-b-gal) staining

HCAECs, pretreated or not pretreated, were incubated withGW501516 for 24 h and then exposed to Ang II for 3 days in six-well plates. After washing twice with ice-cold phosphate-bufferedsaline (PBS), a SA-b-gal assay was performed using a senescentcell staining kit (Sigma–Aldrich) according to the manufacturer’sinstructions. SA-b-gal-positive cells were visualized using anOlympus JP/1X71 fluorescence microscope (Olympus, Tokyo,Japan) with digital camera output and were scored. Stainedcells were counted by an independent observer over six individual

low-power fields and represented as a percentage of the totalnumber of cells.

2.4. Gene silencing with small interfering RNA (siRNA)

Cells were transfected with 80 nM control siRNA (Ambion,Austin, TX, USA), 80 nM human PPARd siRNA (Ambion) or 200 nMsiRNA designed against the human SIRT1 mRNA sequence (50-GATGAA GTT GAC CTC CTC A-30 and 50-TGA AGT GCC TCA GAT ATT A-30)(Samchully Pharm, Seoul, Korea) in serum-free medium usingWelfect-Q (WelGENE, Daegu, Korea). Following incubation for 6 h,the transfection medium was replaced with fresh medium. Afterincubation for 30 h, the cells were exposed to the indicatedreagents for the indicated time, and the effects of gene silencingwere analyzed.

2.5. Western blot analysis

Cells exposed to the indicated reagents were washed in ice-cold PBS and lysed in PRO-PREP protein extraction solution(iNtRON Biotechnology, Seoul, Korea). An aliquot of the cell lysatewas subjected to SDS-polyacrylamide gel electrophoresis andtransferred onto a Hybond-P+ polyvinylidene difluoride mem-brane (Amersham Biosciences, UK Ltd., UK). Membranes blockedwith 5% nonfat milk in Tris-buffered saline (TBS) containing 0.1%Tween-20 overnight at 4 8C were incubated with antibodies in TBScontaining 0.05% Tween-20 and 1% BSA overnight at 4 8C. Themembranes were then incubated with peroxidase-conjugatedsecondary antibody for 2 h at room temperature. After extensivewashing in TBS containing 0.1% Tween-20 and 0.1% BSA,immunoreactive bands were detected using West-ZOL Plus(iNtRON Biotechnology).

2.6. Northern blot analysis

Aliquots of 5 mg of total RNA were heat-denatured at 65 8C for15 min in gel-running buffer (40 mM MOPS, 10 mM sodiumacetate and 1 mM EDTA, pH 7.0) containing 50% formamide andwere electrophoresed on 1% agarose gel containing 2.2 Mformaldehyde. The size-fractionated RNA was transferred onto aHybond-N+ nylon membrane (Amersham Biosciences) overnightby capillary action and immobilized using UV Stratalinker(Stratagene, La Jolla, CA, USA). After hybridization with a 32P-labeled human SIRT1 cDNA probe at 68 8C in QuikHyb solution(Stratagene), the membrane was washed and the radioactivity onthe membrane was detected by a Fuji BAS-2500 BioimagingAnalyzer (Fujifilm, Tokyo, Japan). The blots were stripped andrehybridized with a 32P-labeled GAPDH cDNA probe. The cDNAprobe was generated by PCR using primers specific for nucleotides1711–2230 of human SIRT1.

2.7. Measurement of intracellular ROS

To assess intracellular ROS levels, the fluorescent probe 20,70-dichlorofluorescein diacetate (H2DCF-DA; Calbiochem) was used.Cells were seeded on 35 mm cover-glass bottom dishes (SPL LifeSciences, Seoul, Korea) and exposed to the indicated reagents forthe indicated time. The cells were subsequently incubated with10 mM H2DCF-DA for a final 30 min incubation. After washing withice-cold PBS twice, the cells were immediately analyzed using aFACSCalibur flow cytometer (Becton Dickinson Biosciences, SanJose, CA, USA). DCF fluorescence intensity was measured using anexcitation wavelength of 488 nm and emission wavelength of530 nm. In each sample, approximately 10,000 cells were analyzedusing the CellQuest software (Becton Dickinson Biosciences). Foranalysis of ROS using a laser confocal fluorescence microscope, the

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–1634 1629

cells were treated as described above and green fluorescencecorresponding to the levels of intracellular ROS was detectedthrough a 520 nm long-pass filter using an Olympus FV-1000 laserfluorescence microscope (Olympus).

2.8. Reporter gene assay

The human SIRT1 luciferase reporter vector (pGL4, �2487 to�30) was generously provided by Dr. Masafumi Ito (Department ofLongevity and Aging Research, Kifu International Institute ofBiotechnology, Japan). HCAECs were seeded into 6-well Plates 18–24 h prior to transfection, then co-transfected with 1 mg SIRT1luciferase reporter plasmid and 0.5 mg SV40 b-galactosidaseexpression vector (pSV b-Gal vector, Promega) using SuperFectreagent (Qiagen, Valencia, CA). After incubation for 24 h, cells wereincubated for 3 h in the presence of WY-14643, GW501516,troglitazone, rosiglitazone or DMSO. Cells were lysed in luciferasereporter lysis buffer (Promega) and an aliquot of the total lysateswas used to measure the luciferase activity with Microlumat PlusLB96 V (EG&G Berthold, Bad Wildbad, Germany). Variations intransfection efficiency were normalized using b-galactosidaseactivity.

2.9. Statistical analysis

Data are expressed as means � SE. Statistical significance wasdetermined by Student’s t-test or ANOVA with post hoc Bonferronitest. A value of P < 0.05 was considered statistically significant.

3. Results

3.1. Activation of PPARd inhibits Ang II-induced premature

senescence and ROS generation in HCAECs

Since Ang II has been implicated in vascular aging [27], weassessed whether activation of PPARd might affect prematuresenescence in HCAECs exposed to Ang II. HCAECs exposed to Ang

Fig. 1. Activation of PPARd by GW501516 attenuates Ang II-induced premature senescenc

were incubated in the presence or absence of Ang II for 3 days. Senescent cells were dete

percentage of SA-b-gal-positive cells (B). Bars indicate 100 mm. Representative images

means � SE (n = 4). *P < 0.01 compared to untreated group; #P < 0.01 compared to Ang II-tre

6 h. Intracellular ROS accumulation was analyzed using a FACSCalibur with H2DCF-DA, a per

are shown.

II exhibited a significant increase in SA-b-gal activity, abiomarker for cellular senescence, relative to control cells. Incontrast, the increase in SA-b-gal activity was significantlyreduced in the presence of GW501561, a PPARd specific ligand,demonstrating the involvement of PPARd in the inhibition of AngII-induced premature senescence (Fig. 1A and B). In addition,since Ang II is also known to produce ROS in vascular endothelialcells [28], we examined the effects of GW501516 on Ang II-induced production of ROS in HCAECs. While Ang II increasedROS levels, pretreatment with GW501516 markedly suppressedAng II-induced ROS generation (Fig. 1C). When the cells werepretreated with 100 nM GW501516, the inhibitory effects ofGW501516 on the SA-b-gal activity and ROS generation weredetected at 3 h, which continued up to 48 h of pretreatment (datanot shown).

To clarify the role of PPARd in blocking Ang II-inducedsenescence and ROS generation, the effect of GW501516 wasassessed in HCAECs transfected with a small interfering (si) RNAagainst PPARd. The level of PPARd in HCAECs was markedlyreduced upon transfection with PPARd siRNA in the presence orabsence of GW501516, whereas control siRNA, consisting of a poolof nonspecific sequences, had no effect on PPARd levels (Fig. 2A). Asexpected, the reduction in SA-b-gal activity and ROS generation byGW501516 was recovered in cells transfected with PPARd siRNA,suggesting a PPARd-dependent effect of GW501516 on ROSgeneration and cellular senescence (Fig. 2B–D).

3.2. Activation of PPARd induces SIRT1 mRNA expression in HCAECs

Exposure of HCAECs to GW501516 increased the expression ofSIRT1 transcript in a time- and concentration-dependent manner.Maximum levels were obtained after 3 h of exposure to 100–200 nM GW501516 (Fig. 3A). When cells were exposed to 100 nMGW501516, the increase in SIRT1 mRNA was significant at 1 h andreached a maximum at 3 h (Fig. 3B). An elevated level of SIRT1protein was peaked at 6 h after incubation with GW501516, andcontinued for up to 48 h with slight decline (Fig. 3C).

e and ROS generation in HCAECs. (A and B) Cells pretreated with GW501516 for 24 h

cted by SA-b-gal staining (A) and quantified using image analyzer and plotted as a

from four independent experiments are shown. The results are expressed as the

ated group. (C) Cells pretreated with GW501516 for 24 h were incubated with Ang II for

oxide-sensitive dye, and representative images from four independent determinations

Fig. 2. A small interfering RNA against PPARd abolished the effect of GW501516 on Ang II-induced premature senescence and ROS production in HCAECs. (A) Cells transfected

with PPARd siRNA (80 nM) or control siRNA (80 nM) for 48 h were incubated in the presence or absence of GW501516 (100 nM) for 24 h and then harvested. Cell lysates were

separated by electrophoresis and immunoblotted with anti-PPARd or anti-b-actin antibodies. (B and C) Cells transfected with siRNA against PPARd (80 nM) or control siRNA

(80 nM) were incubated with GW501516 (100 nM) for 24 h and then treated with Ang II (1 mM). After incubation for 6 h, the cells were treated with a peroxide-sensitive dye,

H2DCF-DA (10 mM), during the final 30 min of incubation. The intracellular ROS levels were detected by confocal laser fluorescence microscopy (B) and quantified (C). Bars

indicate 100 mm. Representative images from four independent experiments are shown. (D) Cells treated as described above were incubated with Ang II for 3 days. SA-b-gal

staining was performed and quantified. The results are expressed as the means � SE (n = 4). *P < 0.01 compared to untreated group; #P < 0.01 compared to Ang II-treated group.yP < 0.01, zP < 0.05 compared to Ang II + GW501516-treated group.

Fig. 3. A PPARd ligand induces SIRT1 mRNA expression in HCAECs. (A–C) Primary cultured endothelial cells were incubated for 3 h with various concentrations of GW501516

(A) or exposed to 100 nM GW501516 for the times indicated (B and C). (D) Cells were transfected with PPARd siRNA or control siRNA and harvested at 48 h after transfection.

Northern and Western blot analysis were performed using cDNA probes (A, B and D) and anti-SIRT1 antibody (C), respectively. An image analyzer was used to quantify band

intensity, and the ratio of SIRT1 to GAPDH or b-actin is indicated above each lane. Data are representative results from three or four experiments.

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–16341630

Fig. 4. Activation of PPARd, but not of PPARa or g, induces SIRT1 expression at the transcriptional level. (A) Cells were exposed for 3 h to GW501516 (a specific agonist of

PPARd), WY-14643 (a specific agonist of PPARa), troglitazone (a specific agonist of PPARg), or rosiglitazone (a specific agonist of PPARg). Total protein was extracted and

subjected to Western blot analysis using specific antibodies against each protein. (B) SIRT1 promoter activity was determined in cells transfected with a SIRT1 reporter gene.

Cells transfected with a SIRT1 promoter construct (�2487/�30) were grown for 24 h and then exposed to each PPAR ligand for 3 h. Values are expressed as fold induction

relative to control (means � SE, n = 4). *P < 0.01 compared to untreated group.

Fig. 5. Knockdown of SIRT1 by siRNA abrogated the effects of GW501516 on the

premature senescence and ROS production induced by Ang II in HCAECs. (A) Cells

were transfected with SIRT1 siRNA or control siRNA and harvested at 48 h after

transfection. Cell lysates were separated by electrophoresis and immunoblotted

with anti-SIRT1 or anti-b-actin antibodies. (B) Cells transfected with siRNA against

SIRT1 were incubated with GW501516 for 24 h and then exposed to Ang II for 6 h.

After incubation with 10 mM H2DCF-DA for the final 30 min, the intracellular ROS

levels were detected by confocal laser fluorescence microscopy. (C) Cells treated as

described above were incubated with Ang II for 3 days. SA-b-gal staining was

performed and quantified. The results are expressed as the means � SE (n = 4).

*P < 0.01 compared to untreated group; #P < 0.01 compared to Ang II-treated group.zP < 0.05 compared to Ang II + GW501516-treated group.

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–1634 1631

To further characterize the role of PPARd in the upregulation ofSIRT1 by GW501516, HCAECs were transfected with siRNA againstPPARd. As shown in Fig. 3D, the siRNA-mediated down-regulationof PPARd reversed the expression pattern of SIRT1 induced by thePPARd agonist. These data indicate that SIRT1 expression isregulated in a PPARd-dependent manner.

3.3. Activation of PPARd, but not PPARa or PPARg, induces SIRT1

expression in HCAECs

To identify which PPAR isoforms are involved in the upregula-tion of SIRT1, HCAECs were exposed to specific ligands for PPARd,PPARa and PPARg. HCAECs constitutively express low levels ofSIRT1 protein, and this level increased following activation ofPPARd; however, there was no significant difference in the levelsof SIRT1 protein following activation of either PPARa or PPARg(Fig. 4A).

To determine whether the regulation of SIRT1 expressionthrough PPARd occurs at the level of transcription, a reporter assaywas performed using a luciferase reporter construct driven by thehuman SIRT1 promoter. Activation of PPARd by GW501516significantly enhanced the promoter activity of SIRT1, which isconsistent with the observed increase in the SIRT1 protein level. Incontrast, ligands for either PPARa or PPARg exhibited little or noeffect on SIRT1 promoter activity (Fig. 4B), suggesting that only thePPARd isoform is involved in SIRT1 upregulation.

3.4. Downregulation of SIRT1 abrogates the effect of PPARd against

premature senescence and ROS production

To confirm the role of SIRT1 in the PPARd-mediated inhibitionof cellular senescence and ROS generation induced by Ang II, theexpression of SIRT1 was knocked down by siRNA in HCAECs(Fig. 5A). GW501516 suppressed both Ang II-induced ROSgeneration and premature senescence in HCAECs, whereasknockdown of SIRT1 by siRNA significantly reversed the effect ofGW501516 on both ROS generation and cellular senescence inHCAECs following exposure to Ang II (Fig. 5B and C). These resultsindicate that PPARd inhibits Ang II-induced ROS generation andpremature senescence through SIRT1.

3.5. SIRT1 is essential for the suppression of Ang II-induced ROS

generation and premature senescence by PPARd in HCAECs

To further clarify the functional significance of SIRT1 upregula-tion by PPARd, we assessed the impact of SIRT1 activity on thePPARd-mediated suppression of ROS generation and subsequentcellular senescence induced by Ang II. Inhibition of SIRT1 by

Fig. 6. Regulation of SIRT1 activity by activator or inhibitor mimicked the effects of

PPARd on ROS generation and premature senescence induced by Ang II in HCAECs.

(A) Cells pretreated with GW501516 for 24 h were exposed to sirtinol or resveratrol.

After incubation for 30 min, cells were stimulated with Ang II for 6 h and then

intracellular ROS levels were analyzed by confocal laser fluorescence microscopy

using H2DCF-DA, a peroxide-sensitive dye. (B) Cells treated as described above were

incubated with Ang II for 3 days and then senescent cells were analyzed by SA-b-gal

staining. The results are expressed as the means � SE (n = 4). *P < 0.01 compared to

untreated group; #P < 0.01 compared to Ang II-treated group. yP < 0.01 compared to

Ang II + GW501516-treated group.

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–16341632

sirtinol, an inhibitor of SIRT1, significantly counteracted thesuppression of ROS generation and premature senescence byGW501516 (Fig. 6). In contrast, activation of SIRT1 by resveratrol,an activator of SIRT1, significantly inhibited both Ang II-inducedROS generation and premature senescence to the similar extent asdid GW501516 in HCAECs (Fig. 6). Furthermore, addition ofresveratrol to GW501516-treated cells significantly potentiatedthe inhibitory effects of GW501516, indicating that the anti-senescence or anti-oxidative activity of PPARd is mediated throughSIRT1 (Fig. 6). These results suggest that PPARd regulates Ang II-induced ROS generation and premature senescence by modulatingthe expression of SIRT1.

4. Discussion

A growing body of evidence indicates that SIRT1 is vascular-protective, particularly against the endothelium-dependent regu-lation of vascular homeostasis and premature senescence inducedby oxidative stresses [22,23,29,30]. SIRT1 is a mediator of the effectof calorie restriction and cilostazol on the endothelial-dependentcontrol of vascular tone and premature senescence induced byoxidative stress [23,29]. In fact, the reduced expression and activityof SIRT1 is involved in the acceleration of senescence induced byhigh glucose in endothelial progenitor cells [31]. Although SIRT1 isknown to inhibit cellular senescence in endothelial cells, little isknown about the molecular switch that regulates its activityor expression. The present study found that the levels of ROSand the number of senescent cells were significantly reduced in

Ang II-treated HCAECs in the presence of GW501516, a specificagonist of PPARd. In contrast, SIRT1, a known anti-senescencefactor [22–25], was significantly upregulated in HCAECs exposedto GW501516. siRNA-mediated inhibition of SIRT1 expression orsuppression of SIRT1 activity antagonized the GW501516-medi-ated inhibition of premature senescence and increased ROSproduction, whereas resveratrol, an activator of SIRT1, potentiatedthe effects of GW501516 by inhibiting Ang II-induced prematuresenescence in HCAECs.

The anti-aging properties of PPARd are linked to its induction ofthe intracellular anti-senescence factor SIRT1 in vascular endothe-lium. The present finding is in line with previous investigationshowing that activation of PPARd by a specific agonist inhibitspremature senescence through PTEN-mediated modulation ofPI3K/Akt/Rac1 signaling in human vascular smooth muscle cellsexposed to Ang II [14]. In addition, activation of PPARg, anothermember of the PPAR family, also inhibits angiotensin II-inducedsenescence in endothelial progenitor cells by down-regulating theexpression of the angiotensin II type 1 receptor [32]. In contrast, adifferent line of study showed that activation of PPARg by ligandsaccelerated replicative senescence by upregulating expression ofp16, a cell cycle inhibitor that triggers the onset of cellularsenescence in human fibroblasts [33]. Although the role of thePPAR nuclear receptor family in cellular senescence is controver-sial, the present data clearly indicate that PPARd activation inhibitsAng II-induced premature senescence in HCAECs. Since senescentcells have been detected in human atherosclerosis lesions [5,34], itmay be possible to control the development of age-associatedcardiovascular disease, such as atherosclerosis, through PPARdactivation. Accordingly, the present findings provide new insightsinto the primary role of PPARd as a promising new target fortherapeutic intervention in vascular endothelium.

In aging vasculature, senescence is associated with theaccumulation of oxidative stress in endothelial cells [23,35,36].In line with these findings, the present data demonstrates thatactivation of PPARd by an agonist significantly inhibits Ang II-induced ROS generation in HCAECs. NADPH oxidase has beenimplicated in ROS generation in vascular cells [14,37]. In fact,activation of PPARd suppresses the translocation of Rac1 to theplasma membrane, an important event in the initiation of ROSproduction by NADPH oxidase in VSMCs [37]. Accordingly, ROSproduction by NADPH oxidase seems to play a major role in Ang II-induced senescence in HCAECs. The present study shows thatligand-activated PPARd significantly inhibits ROS production.Although the effects of PPARd on NADPH oxidase were not directlyassessed, our previous study demonstrated that ligand-activatedPPARd counteracts Ang II-induced ROS generation by inhibitingrac1 translocation in VSMCs [37]. SIRT1 was also shown toparticipate in the inhibition of oxidative stress-induced cellularsenescence in human endothelium [23,38]. In fact, upregulation ofSIRT1 by GW501516 inhibits Ang II-induced cellular senescence inHCAECs and resveratrol-mediated activation of SIRT1 potentiatesthe effects of GW501516 on the ROS and premature senescence,although recent reports have been demonstrated that resveratrol isnot a direct activator of SIRT1 [39,40]. However, addition ofresveratrol or sirtinol, known as an activator or inhibitor of SIRT1,respectively, modulates both the ROS generation and prematuresenescence induced by Ang II in our experimental conditions.These findings therefore suggest that PPARd suppresses Ang II-induced oxidative stress by a mechanism involving SIRT1.

PPARd-mediated induction of SIRT1 expression in HCAECs is akey event in the inhibition of cellular senescence induced byangiotensin II. A longevity-related factor, SIRT1, a NAD+-dependenthistone deacetylase, modulates a variety of cellular processes,including cell cycle, senescence, apoptosis and metabolism [17–21].Previous reports indicated that the transcriptional regulation of

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–1634 1633

SIRT1 may be complex and involve energy metabolism-relatedpathways such as caloric restriction in mammals [41–43]. Regardingthe transcriptional regulation of SIRT1, the involvement oftranscription factors, such as BRCA1, E2F1, HIC1 and TLX, has alsobeen demonstrated in the context of tumorigenesis, cell prolifera-tion and apoptosis [44–47]; however, the complete transcriptionalregulation of SIRT1 is poorly understood. PPARd was originallyshown to promote the expression of SIRT1 in human hepatocyte-derived cells [26]. This change in SIRT1 expression is mediated by anon-conventional signaling mechanism in which Sp1, but not PPRE,plays a pivotal role. In contrast, the transcription of SIRT1 is inhibitedby ligand-activated PPARg in senescent cells [48]. Inhibition of SIRT1by PPARg occurs in the context of a negative feedback and self-regulation loop via acetylation of PPARg through the directinteraction of both proteins. Further studies are therefore necessaryto clarify the details of the PPARd/SIRT1 interaction in thetranscriptional regulation of SIRT1 by GW501516.

The present findings indicate that activation of PPARd regulatesAng II-induced cellular senescence in endothelial cells by inducingSIRT1 expression and thereby modulating cellular ROS generation.The effect of PPAR ligands on SIRT1 expression may mediate manyof the functions of PPAR, such as cellular senescence and anti-inflammatory activity, by regulating epigenetic pathways. Accord-ingly, the current data support the hypothesis that PPARd is a keytarget for therapeutic intervention in age-related cardiovasculardisorders.

Acknowledgments

Work was supported in part by a National Research Foundation(NRF) grant funded by the Korean government (2012-0005311),and by the Next-Generation BioGreen 21 Program (No. PJ007980),Rural Development Administration, Republic of Korea.

References

[1] Vanhoutte PM. Endothelial control of vasomotor function: from health tocoronary disease. Circ J 2003;67:572–5.

[2] Freedman DA. Senescence and its bypass in the vascular endothelium. FrontBiosci 2005;10:940–50.

[3] Brandes RP, Fleming I, Busse R. Endothelial aging. Cardiovasc Res 2005;66:286–94.

[4] Erusalimsky JD, Skene C. Mechanisms of endothelial senescence. Exp Physiol2009;94:299–304.

[5] Voghel G, Thorin-Trescases N, Farhat N, Nguyen A, Villeneuve L, MamarbachiAM, et al. Cellular senescence in endothelial cells from atherosclerotic patientsis accelerated by oxidative stress associated with cardiovascular risk factors.Mech Ageing Dev 2007;128:662–71.

[6] Ogryzko VV, Hirai TH, Russanova VR, Barbie DA, Howard BH. Human fibroblastcommitment to a senescence-like state in response to histone deacetylaseinhibitors is cell cycle dependent. Mol Cell Biol 1996;16:5210–8.

[7] Chen Q, Ames BN. Senescence-like growth arrest induced by hydrogen perox-ide in human diploid fibroblast F65 cells. Proc Natl Acad Sci U S A 1994;91:4130–4.

[8] Chen QM, Prowse KR, Tu VC, Purdom S, Linskens MH. Uncoupling the senes-cent phenotype from telomere shortening in hydrogen peroxide-treatedfibroblasts. Exp Cell Res 2001;265:294–303.

[9] Ehrenborg E, Krook A. Regulation of skeletal muscle physiology and metabo-lism by peroxisome proliferator-activated receptor delta. Pharmacol Rev2009;61:373–93.

[10] Jandeleit-Dahm KA, Calkin A, Tikellis C, Thomas M. Direct antiatheroscleroticeffects of PPAR agonists. Curr Opin Lipido 2009;20:24–9.

[11] Lee CH, Chawla A, Urbiztondo N, Liao D, Boisvert WA, Evans RM, et al.Transcriptional repression of atherogenic inflammation: modulation by PPAR-delta. Science 2003;302:453–7.

[12] Takata Y, Liu J, Yin F, Collins AR, Lyon CJ, Lee CH, et al. PPARdelta-mediatedantiinflammatory mechanisms inhibit angiotensin II-accelerated atheroscle-rosis. Proc Natl Acad Sci U S A 2008;105:4277–82.

[13] Kim HJ, Ham SA, Kim SU, Hwang JY, Kim JH, Chang KC, et al. Transforminggrowth factor-beta1 is a molecular target for the peroxisome proliferator-activated receptor delta. Circ Res 2008;102:193–200.

[14] Kim HJ, Ham SA, Kim MY, Hwang JS, Lee H, Kang ES, et al. PPARd coordinatesangiotensin II-induced senescence in vascular smooth muscle cells throughPTEN-mediated inhibition of superoxide generation. J Biol Chem 2011;286:44585–93.

[15] Oliver Jr WR, Shenk JL, Snaith MR, Russell CS, Plunket KD, Bodkin NL, et al.A selective peroxisome proliferator-activated receptor delta agonist pro-motes reverse cholesterol transport. Proc Natl Acad Sci U S A 2001;98:5306–11.

[16] Wang YX, Lee CH, Tiep S, Yu RT, Ham J, Kang H, et al. Peroxisome-proliferator-activated receptor delta activates fat metabolism to prevent obesity. Cell2003;113:159–70.

[17] Michan S, Sinclair D. Sirtuins in mammals: insights into their biologicalfunction. Biochem J 2007;404:1–13.

[18] Cohen HY, Lavu S, Bitterman KJ, Hekking B, Imahiyerobo TA, Miller C, et al.Acetylation of the C terminus of Ku70 by CBP and PCAF controls Bax-mediatedapoptosis. Mol Cell 2004;13:627–38.

[19] Brunet A, Sweeney LB, Sturgill JF, Chua KF, Greer PL, Lin Y, et al. Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase.Science 2004;303:2011–5.

[20] Langley E, Pearson M, Faretta M, Bauer UM, Frye RA, Minucci S, et al. HumanSIR2 deacetylates p53 and antagonizes PML/p53-induced cellular senescence.EMBO J 2002;21:2383–96.

[21] Vaziri H, Dessain SK, Ng Eaton E, Imai SI, Frye RA, Pandita TK, et al. hSIR2(-SIRT1) functions as an NAD-dependent p53 deacetylase. Cell 2001;107:149–59.

[22] Zu Y, Liu L, Lee MY, Xu C, Liang Y, Man RY, et al. SIRT1 promotes proliferationand prevents senescence through targeting LKB1 in primary porcine aorticendothelial cells. Circ Res 2010;106:1384–93.

[23] Ota H, Eto M, Kano MR, Ogawa S, Iijima K, Akishita M, et al. Cilostazolinhibits oxidative stress-induced premature senescence via upregulation ofSirt1 in human endothelial cells. Arterioscler Thromb Vasc Biol 2008;28:1634–9.

[24] Ota H, Eto M, Ogawa S, Iijima K, Akishita M, Ouchi Y. SIRT1/eNOS axis as apotential target against vascular senescence, dysfunction and atherosclerosis.J Atheroscler Thromb 2010;17:431–5.

[25] Miyazaki R, Ichiki T, Hashimoto T, Inanaga K, Imayama I, Sadoshima J, et al.SIRT1, a longevity gene, downregulates angiotensin II type 1 receptor expres-sion in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol 2008;28:1263–9.

[26] Okazaki M, Iwasaki Y, Nishiyama M, Taguchi T, Tsugita M, Nakayama S, et al.PPARbeta/delta regulates the human SIRT1 gene transcription via Sp1. Endocr J2010;57:403–13.

[27] Mukai Y, Shimokawa H, Higashi M, Morikawa K, Matoba T, Hiroki J, et al.Inhibition of renin-angiotensin system ameliorates endothelial dysfunctionassociated with aging in rats. Arterioscler Thromb Vasc Biol 2002;22:1445–50.

[28] Li JM, Shah AM. Endothelial cell superoxide generation: regulation and rele-vance for cardiovascular pathophysiology. Am J Physiol Regul Integr CompPhysiol 2004;287:R1014–30.

[29] Mattagajasingh I, Kim CS, Naqvi A, Yamamori T, Hoffman TA, Jung SB, et al.SIRT1 promotes endothelium-dependent vascular relaxation by activatingendothelial nitric oxide synthase. Proc Natl Acad Sci U S A 2007;104:14855–60.

[30] Potente M, Dimmeler S. Emerging roles of SIRT1 in vascular endothelialhomeostasis. Cell Cycle 2008;7:2117–22 [review].

[31] Balestrieri ML, Rienzo M, Felice F, Rossiello R, Grimaldi V, Milone L, et al. Highglucose downregulates endothelial progenitor cell number via SIRT1. BiochimBiophys Acta 2008;1784:936–45.

[32] Imanishi T, Kobayashi K, Kuroi A, Ikejima H, Akasaka T. Pioglitazone inhibitsangiotensin II-induced senescence of endothelial progenitor cell. HypertensRes 2008;31:757–65.

[33] Gan Q, Huang J, Zhou R, Niu J, Zhu X, Wang J, et al. PPAR{gamma} acceleratescellular senescence by inducing p16INK4{alpha} expression in human diploidfibroblasts. J Cell Sci 2008;121:2235–45.

[34] Minamino T, Miyauchi H, Yoshida T, Ishida Y, Yoshida H, Komuro I. Endothelialcell senescence in human atherosclerosis: role of telomere in endothelialdysfunction. Circulation 2002;105:1541–4.

[35] Oeseburg H, Iusuf D, van der Harst P, van Gilst WH, Henning RH, Roks AJ.Bradykinin protects against oxidative stress-induced endothelial cell senes-cence. Hypertension 2009;53:417–22.

[36] Zhan H, Suzuki T, Aizawa K, Miyagawa K, Nagai R. Ataxia telangiectasiamutated (ATM)-mediated DNA damage response in oxidative stress-inducedvascular endothelial cell senescence. J Biol Chem 2010;285:29662–70.

[37] Lee H, Ham SA, Kim MY, Kim JH, Paek KS, Kang ES, et al. Activation of PPARdcounteracts angiotensin II-induced ROS generation by inhibiting rac1translocation in vascular smooth muscle cells. Free Radic Res 2012;46:912–9.

[38] Kao CL, Chen LK, Chang YL, Yung MC, Hsu CC, Chen YC, et al. Resveratrolprotects human endothelium from H(2)O(2)-induced oxidative stress andsenescence via SirT1 activation. J Atheroscler Thromb 2010;17:970–9.

[39] Boher D, Wu J, Cumine S, Kim KW, Lu SC, Atangan L, et al. Resveratrol is not adirect activator of SIRT1 enzyme activity. Chem Biol Drug Des 2009;74:619–24.

[40] Pacholec M, Bleasdale JE, Chrunyk B, Cunningham D, Flynn D, Garofalo RS, et al.SRT1720, SRT2183, SRT1460, and resveratrol are not direct activators of SIRT1.J Biol Chem 2010;285:8340–51.

[41] Cohen HY, Miller C, Bitterman KJ, Wall NR, Hekking B, Kessler B, et al. Calorierestriction promotes mammalian cell survival by inducing the SIRT1 deace-tylase. Science 2004;305:390–2.

[42] Chen D, Steele AD, Lindquist S, Guarente L. Increase in activity during calorierestriction requires Sirt1. Science 2005;310:1641.

M.Y. Kim et al. / Biochemical Pharmacology 84 (2012) 1627–16341634

[43] Boily G, Seifert EL, Bevilacqua L, He XH, Sabourin G, Estey C, et al. SirT1regulates energy metabolism and response to caloric restriction in mice. PLoSOne 2008;3:e1759.

[44] Wang RH, Zheng Y, Kim HS, Xu X, Cao L, Luhasen T, et al. Interplay amongBRCA1, SIRT1, and Survivin during BRCA1-associated tumorigenesis. Mol Cell2008;32:11–20.

[45] Chen WY, Wang DH, Yen RC, Luo J, Gu W, Baylin SB. Tumor suppressor HIC1directly regulates SIRT1 to modulate p53-dependent DNA-damage responses.Cell 2005;123:437–48.

[46] Wang C, Chen L, Hou X, Li Z, Kabra N, Ma Y, et al. Interactions between E2F1 andSirT1 regulate apoptotic response to DNA damage. Nat Cell Biol 2006;8:1025–31.

[47] Iwahara N, Hisahara S, Hayashi T, Horio Y. Transcriptional activation of NAD+-dependent protein deacetylase SIRT1 by nuclear receptor TLX. Biochem Bio-phys Res Commun 2009;386:671–5.

[48] Han L, Zhou R, Niu J, McNutt MA, Wang P, Tong T. SIRT1 is regulated by aPPAR{g}-SIRT1 negative feedback loop associated with senescence. NucleicAcids Res 2010;38:7458–71.