lable at ScienceDirect

Atherosclerosis 235 (2014) 247e255

Contents lists avai

Atherosclerosis

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

Lack of vitamin D receptor causes stress-induced prematuresenescence in vascular smooth muscle cells through enhanced localangiotensin-II signals

Petya Valcheva a, Anna Cardus a, Sara Panizo a, Eva Parisi a, Milica Bozic a,Jose M. Lopez Novoa c, Adriana Dusso a, Elvira Fernández b,1, Jose M. Valdivielso a,*,1

a Experimental Nephrology Laboratory, Department of Experimental Medicine, Biomedical Research Institute of Lleida (IRBLLEIDA), Lleida, SpainbNephrology Service and UDETMA, University Hospital Arnau de Vilanova, Lleida, SpaincDepartment of Physiology and Pharmacology, University of Salamanca, Salamanca, Spain

a r t i c l e i n f o

Article history:Received 17 February 2014Received in revised form9 April 2014Accepted 1 May 2014Available online 13 May 2014

Keywords:VSMCVitamin DAngiotensin-IIROSSenescence

* Corresponding author. Experimental NephroloResearch Institute (IRBLLEIDA), Edificio BiomedicinRoure 80, 25198 Lleida, Spain. Tel.: þ34 973003650;

E-mail address: valdivielso@medicina.udl.es (J.M.1 EF and JMV share senior authorship.

http://dx.doi.org/10.1016/j.atherosclerosis.2014.05.9110021-9150/� 2014 Elsevier Ireland Ltd. All rights rese

a b s t r a c t

Objectives: The inhibition of the renal renin-angiotensin system by the active form of vitamin D con-tributes to the cardiovascular health benefits of a normal vitamin D status. Local production ofangiotensin-II in the vascular wall is a potent mediator of oxidative stress, prompting prematuresenescence. Herein, our objective was to examine the impact of defective vitamin D signalling on localangiotensin-II levels and arterial health.Methods: Primary cultures of aortic vascular smooth muscle cells (VSMC) from wild-type and vitamin Dreceptor-knockout (VDRKO) mice were used for the assessment of cell growth, angiotensin-II and su-peroxide anion production and expression levels of cathepsin D, angiotensin-II type 1 receptor andp57Kip2. The in vitro findings were confirmed histologically in aortas from wild-type and VDRKO mice.Results: VSMC from VDRKO mice produced more angiotensin-II in culture, and elicited higher levels ofcathepsin D, an enzyme with renin-like activity, and angiotensin-II type 1 receptor, than wild-type mice.Accordingly, VDRKO VSMC showed higher intracellular superoxide anion production, which could besuppressed by cathepsin D, angiotensin-II type 1 receptor or NADPH oxidase antagonists. VDRKO cellspresented higher levels of p57Kip2, impaired proliferation and premature senescence, all of them bluntedupon inhibition of angiotensin-II signalling. In vivo studies confirmed higher levels of cathepsin D,angiotensin-II type 1 receptor and p57Kip2 in aortas from VDRKO mice.Conclusion: The beneficial effects of active vitamin D in vascular health could be a result of the atten-uation of local production of angiotensin-II and downstream free radicals, thus preventing the prematuresenescence of VSMC.

� 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Adverse cardiovascular events are the leading cause of deathamong chronic kidney disease (CKD) and end-stage-renal disease(ESRD) patients. The higher incidence of atherosclerosis in thesepopulations is a well-recognized contributor to their dispropor-tionally high morbi-mortality. In the course of CKD, severalmechanisms increase the risk for cardiovascular disease. The

gy Laboratory, Biomedicala 1, Despacho d1.12, Rovirafax: þ34 973702213.Valdivielso).

rved.

decrease in renal parenchyma function causes a deregulatedactivation of the renin-angiotensin system (RAS). The mainphysiological role of angiotensin-II (Ang-II) is the regulation ofblood pressure, electrolyte balance and extracellular volume, butit is also involved in some pathological conditions includingfibrosis, hypertrophy, cell proliferation and senescence [1]. Ang-IIproduction could be initiated locally through the catalytic ac-tivity of renin or the renin-like enzyme cathepsin D (CatD),depending on the cell type. Local Ang-II binding to its main re-ceptor, Ang-II type 1 receptor (AT1), exerts pro-inflammatoryeffects in the vascular wall through the induction of reactiveoxygen species (ROS) and inflammatory cytokines [2]. Free rad-icals impair vascular cell life-span through the onset of cellularsenescence [3], a common feature of vascular smooth muscle

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255248

cells (VSMC) that accumulate in the atherosclerotic plaqueenhancing its susceptibility to rupture [4]. Thus, the pathophys-iological impact of local Ang-II production on vascular wallhealth deserves further investigation.

An important feature of CKD is also vitamin D (VD) deficiency.The inverse relationship between VD deficiency and cardiovas-cular health is well documented. However, the underlyingmechanisms have not been fully elucidated [5,6]. One of thepotential benefits of the active form of VD (1,25(OH)2D3) is thedown-regulation of renin levels. Li et al. demonstrated higherAng-II plasma levels in VD receptor-knockout (VDRKO) mice, dueto increased renin expression and activity [7]. In vitro experi-ments showed that 1,25(OH)2D3 directly suppresses renin geneexpression by targeting the cyclic-AMP response element (CRE)in the renin gene promoter [8].

The role of VD signalling in VSMC proliferation is controversial.Thus, whereas some studies have shown decreases in proliferation[9e11], we and others have found the opposite effect [12e14]. Thus,previous studies from our laboratory have shown that 1,25(OH)2D3increases proliferation and calcification of VSMC through a VDresponse element in the VEGF promoter [15,16] and via a RANK-BMP4 dependent pathway [17,18], respectively. These are twocritical processes in the vasculature during atherosclerosisprogression.

The present study examines the contribution of the lack of VDsignalling in VSMC to enhance local Ang-II synthesis, increasedfree radical production and VSMC senescence, all of which mayaccelerate the onset of subclinical cardiovascular disease, inde-pendently of the degree of systemic RAS activation. Furthermore, itclarifies the role of vitamin D signalling in the regulation VSMCproliferation.

2. Methods

2.1. Experimental animals

VDRKO mice from the Tokyo strain [19], were kind gift fromDr. Shigeaki Kato (University of Tokyo, Japan). For PCR genotyp-ing four primers were used to amplify a 140 bp VDR band and a455-bp Neo band from the inserted targeting vector, as describedbefore [20]. In order to normalize mineral homoeostasis andexclude hypocalcaemia as a confounding factor in our experi-ments, after weaning only VDRKO mice were fed a rescue diet[20% lactose, 2.0%Ca, 1.25%P and (Harlan Laboratories)]. Three- tosix-months-old male and female VDRKO mice and their WT lit-termates, as a control group, were used for all experiments,following the National Institute of Health Guide for the Care andUse of Laboratory Animals. The protocols used in this study wereapproved by the Ethic Animal Experimentation Committee ofUniversity of Lleida.

2.2. Cells and culture conditions

Aortic VSMC from 3-months-old mice were obtained by ex-plants culture as previously described [15]. Cells were grown in 10%heat-inactivated FBS-containing DMEM (GIBCO) and maintained at37 �C in humidified incubator (5%CO2/95%air). Cells at passages 3e8were serum-starved in 0.2%FBS DMEM during 24 h to synchronizethe population in G0 phase of the cell cycle before treatments.Pepstatin A (PstA, Santa Cruz Biotechnology), inhibitor of asparticproteases cathepsin D, pepsin and renin, was dissolved in water-free DMSO and added at 2.5 mg/ml. Losartan (LOS, SigmaeAldrich), an AT1 antagonist, was dissolved in water and used at10 mmol/L. Diphenyleneiodonium [DPI, kind gift from Dr JordiBoada (University of Lleida, Spain)], a NADPH oxidase inhibitor was

dissolved in DMSO and used at 10 mmol/L. Sodium 4,5-dihydroxybenzene-1,3-disulfonate [Tiron, kind gift from Dr XavierDolcet (University of Lleida, Spain)], a scavenger of superoxideanions, was used at 10 mmol/L 1,25-dihydroxyvitamin D3 (1,25D3,SigmaeAldrich) was dissolved in absolute ethanol and used at100 nmol/L.

2.3. RNA expression

Total cellular RNA was extracted using TRIzol�Reagent (Invi-trogen) following the manufacturer’s instructions. The reversetranscription was performed using a First Strand cDNA SynthesisKit for RT-PCR (AMV) (Roche Diagnostics). For the Real-Time PCRTaqMan� probes for Cathepsin D (CatD, Mn00515587_m1),angiotensin-II receptor type 1a (AT1a, Mm01166161_m1), p57Kip2

(Mm00438170_m1) and glyceraldehyde-3-phosphate dehydroge-nase (GAPDH, Mm99999915_g1), TaqMan� Universal PCR MasterMix, No AmpErase� UNG and ABI PRISM� 7000 sequence detectionPCR system from Applied Biosystems were used following themanufacturer’s instructions. The relative RNA amount was calcu-lated by standard formulae (DDCt method) using GAPDH as ahousekeeping gene.

2.4. Western blot

Cells and arterial tissues were collected in lysis buffer [2%sodium dodecyl sulphate (SDS), 125 mmol/L Tris, pH6.8, 1� prote-ase inhibitor cocktail (SigmaeAldrich), 2 mmol/L phenyl-methanesulfonylfluoride (PMSF)]. Protein concentrations weredetermined with DC� Protein Assay kit (Bio-Rad). Western blotwas performed as previously described [21]. Blots were incubatedwith primary antibody for CatD (1:5000, Abcam), AT1 (1:1000,Santa Cruz Biotechnology), p57Kip2 (1:500, Santa Cruz Biotech-nology), and a-tubulin (1:10000, SigmaeAldrich) overnight at 4 �C.After washing, appropriate horseradish peroxidase�conjugatedsecondary antibody at 1:10,000 dilution (anti-rabbit: Cell SignalingTechnology� and anti-mouse: Jackson ImmunoResearch Labora-tories, Inc.) was added for an extra hour. Blots were developed withthe Amersham ECL� Advance Western Blotting Detection Kit (GEHealthcare Bio-Sciences AB) or EZ-ECL Chemiluminescence detec-tion kit for HRP (Biological Industries) and the VersaDoc Imagingsystem Model 4000 (Bio-Rad).

2.5. Angiotensin-II determination

Ang-II concentration in the culture mediumwas determined byEIA [Angiotensin-II SPIE-IA kit (BertinPharma)], according to themanufacturer’s instructions. The concentration of Ang-II wasnormalized to cell number.

2.6. Superoxide anion detection

For detection of intracellular superoxide anions, after 24 h in10%FBS DMEM, cells were detached and re-suspended in PBS alone,with PstA, LOS or DPI and kept at 37 �C for 30 min. During the last20 min of incubation, 10 mmol/L dihydroethidium (DHE, SigmaeAldrich) was added. Then, the cells were transferred into 5 mlpolystyrene round bottom tubes (BD Biosciences) and analyzed inFACSCantoTMII flow cytometer (BD Biosciences), using FACSDivaVersion 6.1.1 software.

2.7. Senescence-associated-b-galactosidase (SA-b-Gal) activity

SA-b-Gal activity of the cells was assessed, as described pre-viously [22], with minor modifications. Briefly, cells were fixed

Fig. 1. Depletion of VDR enhances angiotensin-II signalling pathway. Expression profiles of Cathepsin D (CatD) mRNA (a) and protein (b) levels in WT cells treated with or without1,25-dihydroxyvitamin D3 at 100 nM and VDRKO cells after 24 h in 0.2%FBS DMEM. (c) Angiotensin-II (Ang-II) concentrations in medium of quiescent WT and VDRKO cells. Ang-IItype 1 receptor (AT1) mRNA (d) and protein (e) levels. The graphs in panels (b) and (e) represent optical density (OD) of protein to OD of a-tubulin, as a loading control. Barsrepresent mean � SD of at least three independent experiments. Statistical significance, *: p < 0.05 and **: p < 0.01 versus WT.

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255 249

during 4 min at RT in 0.5% glutaraldehyde and rinsed in PBS(pH7.2, supplemented with 1 mmol/L MgCl2). Finally, the cellswere incubated overnight at 37 �C (without CO2) with a freshly-prepared staining solution for SA-b-Gal: 1 mg/ml 5-bromo-4-chloro-3-indolyl b-D-galactoside (SigmaeAldrich), 5 mmol/Lpotassium ferrocyanide, 5 mmol/L potassium ferricyanide,2 mmol/L MgCl2 in PBS, pH6.0. In order to count the total cellnumber, the cell nuclei were further stained with Hoechst 33258dye (SigmaeAldrich) during 10 min. The number of SA-b-Galpositive cells and the total cell number were counted anddocumented by a digital imaging system (Nikon). In order toassess the impact of RAS over the appearance of prematuresenescence, we treated the cells with PstA (Santa Cruz Biotech-nology) or LOS (SigmaeAldrich) during 14 days before the SA-b-Gal activity assessment.

2.8. Growth curves

The growth patterns of VSMC were studied in 10%FBS-sup-plemented DMEM. Cells (10 � 103 cells/cm2) were left toattach overnight and counted in Neubauer chamber at 0, 3, 6, 18,24, 32 and 48 h. For the experiments with the AT1 antagonist,cell number was counted 72 h after LOS (SigmaeAldrich)addition.

2.9. Cell cycle fluorescence-activated cell sorter analysis

Cells (10 � 103 cells/cm2) were cultured in 10%FBS DMEM andallowed to reach 50% of confluence, establishing this point as time0. After 48 h the cells were trypsinized and fixed with 70% ethanolovernight at 4 �C. Then, the cells were incubated with 1 mg/ml

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255250

RNAase for 30 min at 37 �C, and afterwards cell nuclei were stainedwith 50 mg/ml propidium iodide during 30 min. Cell fluorescencedistributions were obtained using Coulter Epics XL flow cytometer(Beckman Coulter).

2.10. Immunohistochemistry

Arteries were extracted from isoflurane-anesthesized 6-months-old mice and fixed overnight in 20% formalin solution.After paraffin inclusion, 5 mm thick slices were used for immuno-histochemistry as previously described [23]. Sections were incu-bated with the following primary antibodies: anti-CatD (1:100,Abcam), anti-AT1 (1:400, Santa Cruz Biotechnology) and anti-p57Kip2 (1:100, Santa Cruz Biotechnology). After washings, thesections were incubated with biotinylated secondary antibody(1:50) and stainedwith VECTASTAIN�ABC and peroxidase substrate(DAB) kits (Vector Laboratories) according to manufacturer’s in-structions. Haematoxylin (EnVision� FLEX, Dako) was used tocounterstain cell nuclei.

2.11. Statistical analysis

Data are reported as mean � standard deviation (SD). Statis-tical analysis was performed using Graph Pad Prism (release 5).Results were evaluated by t-test or by two-way analysis of

Fig. 2. VDRKO cells show higher intracellular superoxide anion levels and premature senperoxide anions levels measured by dehydroethidium (DHE) in VSMC lacking vitamin D receproducts, expressed as mean fluorescence units (MFI). ROS production was assessed after thphotographs of WT and VDRKO cells, stained for senescence-associated-b-galactosidase (SAfrom four independent experiments). Bars represent mean � SD of at least three independunder identical experimental conditions.

variance (ANOVA) followed by Bonferroni’s post hoc tests asappropriate. A value of p < 0.05 was considered to denote sta-tistical significance.

3. Results

3.1. Lack of VDR increases local Ang-II signals

In order to provide conclusive evidence for the role of VD invascular health, we examined the expression of several factorsinvolved in Ang-II signalling in primary cultures of aortic VSMC,isolated from WT and VDRKO mice. There was no detectablerenin expression in these cells. However, VDRKO cells showed anincrease in mRNA levels of the acid protease CatD (Fig. 1a).Moreover, when we treated WT cells with 100 nM 1,25(OH)2D3during 24 h, we found a significant decrease in the CatD mRNAlevels. The fact that CatD could be regulated negatively upon VDRactivation, led us focus on CatD-downstream signalling path-ways. The protein levels of CatD were higher in VSMC lacking theVDR (Fig. 1b). Ang-II levels in the cell culture media, measured byEIA, were significantly increased in VDRKO cells (Fig. 1c).Furthermore, AT1 expression was also up-regulated in VDRKOcells at both mRNA (Fig. 1d) and protein levels (Fig. 1e). Theseresults suggest that the absence of VDR signals activates localRAS in VSMC.

escence. (a) Representative flow cytometry histograms showing the intracellular su-ptor (VDRKO) and WT controls from passage 5. (b) Quantification of the DHE oxidatione addition of PstA (2.5 mg/mL), LOS (10 mmol/L) and DPI (10 mmol/L). (c) Representative-b-Gal) activity at pH6.0. Quantification is shown on the right (n ¼ 16, quadruplicatesent experiments. Statistical significance, ns: non significant,***: p < 0.001 vs. WT cells

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255 251

3.2. VDRKO VSMC cells present higher superoxide anion levels andelicit features of premature senescence

Systemic RAS activation is known to increase ROS production inVSMC [24]. To determine local ROS production, the content of su-peroxide anions was measured by flow cytometry in WT andVDRKO cells. Dihydroethidium-derived oxidation products werehigher in VDRKO cells, indicating a higher production of superoxideanions (Fig. 2a). To examine whether the enhancement in freeradicals content in VDRKO cells resulted from an increase in Ang-II-induced NADPH oxidase activity, WT and VDRKO cells were incu-bated with Pepstatin A (PstA, an aspartyl protease inhibitor), Los-artan (LOS, an AT1 antagonist), or diphenyleneiodonium (DPI, aNADPH oxidase inhibitor). As shown in Fig. 2b, all inhibitorssignificantly decreased the superoxide anion content in VDRKOcells to levels similar to those observed in WT controls. The findingthat VSMC lacking VDR grow under a higher oxidative stress couldalso explain their distinct morphology in culture. Specifically,VDRKO cells have a bigger size, a flattened star-like shape and lotsof vacuoles in their cytosol (data not shown), all features of cellsenescence. Staining for Senescence-Associated-b-Galactosidase(SA-b-Gal) activity showed that 22.5% of the VSMC obtained fromVDRKO animals presented a blue perinuclear staining. Instead, WTcells were all negative (Fig. 2c), confirming the appearance ofpremature senescence in VDRKO cells.

3.3. Decreased proliferation rates in VDR lacking VSMC

WT VSMC grew faster than VDRKO, as assessed by the com-parison of the growth curves slopes (p< 0.0001). WTcells followeda typical exponential curve, while VDRKO cells showed a slower,almost linear growth (Fig. 3a). Moreover, FACS analysis alsorevealed a reduction in the percentage of cells in the S-phase of thecell cycle. VDRKO cells presented 73.0 � 4.3% in G1-phase and3.9 � 0.3% in S-phase after 48 h in culture compared to 64.2 � 2.8%

Fig. 3. VDRKO vascular smooth muscle cells show decreased proliferation rates in vitro. (a) GFBS (n ¼ 3, experiments performed in triplicate). (b) FACS analysis to quantify WT and VDRKexpression in WT and VDRKO VSMC. The graph represents optical density (OD) of p57Kip2

cance,*: p < 0.05, ***: p < 0.001 and ns ¼ non significant versus WT.

and 15.4 � 0.8%, respectively for WT cells (Fig. 3b). Importantly, thepercentage of dead cells was similar for both cell types and below5%. To further elucidate the downstream signalling mechanism bywhich VDR depletion induced cell cycle arrest and senescence, theexpression levels of cell cycle checkpoint proteins involved in se-nescent phenotype were examined. An interesting finding was thevery high expression of the cyclin-dependent kinase inhibitor(CDKi) p57Kip2 in VDRKO cells (Fig. 3c).

3.4. Increased local production of angiotensin-II induces freeradicals generation and p57Kip2 upregulation thus contributing todecreased proliferation of VSMC from VDRKO mice

In order to check whether AT1 activation contributed to differ-ences in proliferation rates, LOS exposurewas prolonged to 72 h. LOStreatment increased cell proliferation in VDRKO cells. In contrast, inWT cells, AT1 blockade led to decreased proliferation rates (Fig. 4a).Next, to examine the role of Ang-II signals in the onset of prematuresenescence, VDRKO cells were cultured with PstA or LOS for 14 days.As shown in Fig. 4b, the prolonged exposure to these inhibitorsprevented further increase in the senescence induced by the lack ofVD signalling. None of the WT cells were stained for SA-b-Gal evenupon the treatments. In view of the marked increases in p57Kip2

observed in VDRKO cells, we examined its role in the recovery of theproliferation rates of VDRKO cells after CatD or AT1 inhibition. Fig. 4cshows a significant reduction in the mRNA levels of the CDKi p57Kip2

upon CatD and AT1 blockage. Also the superoxide anion scavengerTiron caused a similar decrease in p57kip2 mRNA levels in VDRKOcells (Fig. 4c) as CatD and AT1 inhibitors.

3.5. VDRKO mice present increased aortic expression of keycomponents of the angiotensin-II signalling cascade

To validate in vivo the link between the lack of VDRand enhanced Ang-II signalling, western blot and

rowth curves of WT (solid line) or VDRKO cells (dotted line) after stimulation with 10%O cell distribution throughout the cell cycle. (c) Representative western blot of p57Kip2

to OD of a-tubulin, as a loading control. Bars represent mean � SD. Statistical signifi-

Fig. 4. The inhibition of the angiotensin-II signaling cascade improves the proliferativecapacity of VDR-ablated VSMC. (a) Proliferation rates of WT and VDRKO VSMC exposedto the AT1 inhibitor Losartan (LOS, 10 mmol/L) for 72 h. (b) Effect of PstA (2.5 mg/mL)and LOS (10 mmol/L) in VDRKO VSMC for 14 days on cell senescence, assessed by SA-b-Gal staining. The percentage of SA-b-Gal positive cells at day 0 was used as a control.(c) mRNA expression of the cell cycle inhibitor p57Kip2 in VDRKO VSMC incubated withPstA (2.5 mg/mL), LOS (10 mmol/L) or Tiron (10 mmol/L) during 24 h (n ¼ 3). Barsrepresent mean � SD. Statistical significance,*: p < 0.05, **: p < 0.01, ***: p < 0.001 andns ¼ non significant versus the corresponding untreated control.

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255252

immunohistochemical (IHC) analysis compared aortic samplesfrom 6-months-old WT and VDRKO mice. The expression of CatD,AT1 and p57Kip2 were markedly higher in the VDRKO arteries(Fig. 5a). Similar findings were observed in IHC analysis (Fig. 5b).

4. Discussion

The present study shows for the first time that the lack of VDsignalling in VSMC induces an increase in local production of Ang-II, which leads to two processes associated with vascular ageing:oxidative stress and premature senescence. Accelerated vascularageing is a characteristic of CKD patients. Even in children withend-stage renal disease, the artery wall presents structural abnor-malities comparable with those found in adults [25]. The VD defi-ciency that characterizes CKD is known to compromisecardiovascular health in these patients. Indeed, the active form ofVD, 1,25(OH)2D3, exerts pro-survival and anti-ageing actions. Theactivated VDR up-regulates Klotho (a gene with numerous anti-ageing properties) through multiple VD response elements in theKlotho promoter [26]. Moreover, Eelen et al. [27] have recentlydemonstrated that VD induces FOXO3a and SESN1 expression inosteoblasts. These are highly conserved proteins that protectagainst age-related pathologies, altering ROS levels. Furthermore,Kim et al. [28] have shown that activation of peroxisomeproliferator-activated receptor b (PPARb), a protein up-regulated byVD [29], controls Ang-II-induced premature senescence in endo-thelial cells by increasing the expression of the novel anti-ageingprotein SIRT1, thereby modulating cellular ROS generation. Like-wise, the fact that activated VDR decreases renin expression [7]may also play a role in the potential beneficial effects of VD oncardiovascular health.

Thus, vascular protection afforded by VD in VSMC could bemediated through several mechanisms, among which the localAng-II pathway has been poorly investigated. It has been describedthat VSMC locally produce Ang-II. The rate-limiting enzyme in renalproduction of Ang-II is renin. We did not detect renin expression inmouse VSMC in vitro nor in vascular tissue in vivo. Fukuda et al. alsowere not able to detect renin in rat VSMC, but they showed that inVSMC obtained from spontaneously hypertensive rats, the higherproduction of Ang-II in vitro was due to increased expression ofCatD [30]. CatD is a lysosomal aspartyl protease with high sequencehomology to renin, which displays renin-like enzymatic activities[31]. In the present study we show for the first time that in theabsence of VDR, VSMC express higher levels of CatD that couldexplain the higher amount of Ang-II. Yuan et al. have demonstratedthat VD blocks the activity of cAMP-response element (CRE) in therenin gene promoter [8]. Interestingly, the promoter of CatD elicitsa putative CRE [32]. Indeed, in our experiments, the incubationwith1,25(OH)2D3 decreased significantly CatD gene expression in WTVSMC. Therefore, CatD expression could be subjected to the samenegative regulation by 1,25(OH)2D3 as the renin gene.

Upon activation of its receptor AT1, Ang-II upregulates NADPHoxidase activity and consequently the production of ROS [33]. Inour study, VDRKO VSMC showed a significant up-regulation of AT1mRNA and protein levels, which could be mediated by increasedAng-II levels, as previously reported [34]. Furthermore, these cellselicited higher ROS levels, which were decreased to levels undis-tinguishable from those in WT after the inhibition of CatD activity,AT1 or NADPH oxidase downstream signalling. Thus, these resultssuggest that the lack of negative control by VD over CatD expres-sion could cause higher Ang-II production with subsequent AT1-mediated NADPH oxidase activation and ROS production.

Ang-II induces AT1-mediated premature senescence in VSMCin vitro through an increased superoxide anion production byNADPH oxidase activity [35]. VSMC lacking VDR showed reducedproliferation and features of cell senescence. Ang-II usuallycontributes to enhanced VSMC growth which could be inhibitedby AT1 receptor antagonists [36], a feature that we observed inWT cells. However, it is possible that in VDRKO cells, LOSincreased proliferation rates as a result of the blockade of Ang-II/

Fig. 5. Higher expression of Ang-II signalling cascade componentseCatD, AT1, and the cell cycle inhibitor p57Kip2 in aorta of VDRKO mice. (a) Representative western blot of CatD,AT1 and p57Kip2 in WT and VDRKO aortic tissue (n ¼ 3). The graph on the right represent optical density (OD) of protein to OD of a-tubulin, as a loading control (b) Immuno-histochemistry for CatD, AT1 and p57Kip2 in aortic samples fromWT and VDRKO mice (n ¼ 3). Scale bars ¼ 100 mm. Bars represent mean � SD Statistical significance, *: p < 0.05 and**: p < 0.01 versus WT.

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255 253

NADPH oxidase-mediated ROS generation. Oxidative stress pro-duces DNA damage, inducing cell cycle inhibitors that make cellsexit the cell cycle. The lower proliferation capacity and theappearance of premature senescence in VDRKO cells could resultfrom the higher expression of the cell cycle inhibitor p57Kip2,implicated in the senescent phenotype [37]. The fact that PstA,LOS and Tiron successfully decreased p57Kip2 levels providesadditional evidence for the contribution of CatD-mediated localRAS activation to the lower proliferation capacity and prematuresenescence of VSMC lacking VDR. Moreover, the inhibition ofCatD activity and AT1-downstream signalling prevented newVDRKO cells to become senescent. Hence, the link between theabsence of functional VDR and cellular ageing could be theexcessive activation of local RAS.

The obvious premature ageing phenotype of VDRKO mice couldbe associated with the increased CatD, AT1 and p57Kip2 expression,found in their arteries. Undoubtedly, these increases could be partlyaccounted for the increased systemic RAS activation, demonstratedby Li et al. [7]. Furthermore, increased local production of Ang-IIcould partly account for the high blood pressure levels describedin these animals. In vivo, VSMC are in quiescent state under normalconditions [38] and their re-entry into the cell cycle occurs invarious pathological situations. For instance, in case of damage,VSMC are stimulated to proliferate in order to repair the vessel.However, the uncontrolled proliferation of VSMC has been linked todiseases such as hypertension and atherosclerosis [39].

The results presented here suggest that the cardiovascularprotection showed by active vitamin D treatment in human CKD

VDRWT VDRKO

1,25(OH)2D3 1,25(OH)2D3

VDR VDR

Cathepsin D Cathepsin D

SENESCENCE

Angiotensinogen

Angiotensin I

Angiotensin II

NADPH oxidase

ROS

p57Kip2

AT1

Losartan

Pepstatin A

DPI

Tiron

Fig. 6. Possible mechanism for the enhanced premature senescence in vascular smooth muscle cells from VDRKO mice. Activation of vitamin D receptor (VDR) by 1,25-dihydroxyvitamin D3 (Vit D) inhibits the cathepsin D gene expression. Therefore, lack of vitamin D signaling results in increased cathepsin D levels, which in turn augmentsangiotensin-II production. The binding of angiotensin-II to its receptor AT1 increases NADPH oxidase activity and free radical production. The latter induces higher p57Kip2 levels andonset of premature senescence.

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255254

patients could be related, at least in part, to the effect of VDRactivation on vascular senescence. Senescent VSMC cells have beenfound in atherosclerotic plaques [3,40], and recent results suggestthat VSMC senescence could even promote atherosclerosis [41].Thus, the accelerated atherosclerotic process that CKD patientssuffer [42,43] could be partially enhanced by the chronic hypo-vitaminosis D associated with CKD. Indeed, recent results haveshown that deletion of VDR gene in an animal model of athero-sclerosis enhances plaque formation [44]. This effect could bemediated, at least in part, by the increase in VSMC senescence andalso by the increase in blood pressure induced both by systemic andlocally produced Ang-II. Although tempting, it is difficult toextrapolate animal data into humans, and experimental resultsshould be taken cautiously until confirmed in clinical settings.

5. Conclusion

The findings of our work suggest that in absence of VD signallingan increase in the local production of Ang-II accounts for theappearance of premature vascular senescence (Fig 6), confirmingthat VD is an essential hormone in the regulation of VSMC prolif-eration. Thus, vitamin D could play a role in both, the normal repairof the vascular wall and in the prevention not only of systemic butalso of local RAS-mediated VSMC hyperproliferation.

Disclosures

No conflicts of interest are declared by the authors.

Acknowledgements

This work was partially supported by grants from FIS PI12/01770, CEDAR and REDINREN (RD12/0021/0026). We thank Dr.Marti Aldea for lively discussions. Drs.Valcheva’s and Cardus’ workto complete this manuscript was partly supported by their

scholarships from ERA-EDTA awarded on November 2011 and July2012, respectively.

References

[1] Dzau VJ. Theodore Cooper Lecture: tissue angiotensin and pathobiology ofvascular disease: a unifying hypothesis. Hypertension 2001;37:1047e52.

[2] Brasier AR, Recinos III A, Eledrisi MS. Vascular inflammation and the renin-angiotensin system. Arterioscler Thromb Vasc Biol 2002;22:1257e66.

[3] Matthews C, Gorenne I, Scott S, Figg N, Kirkpatrick P, Ritchie A, et al.Vascular smooth muscle cells undergo telomere-based senescence in humanatherosclerosis - effects of telomerase and oxidative stress. Circ Res2006;99:156e64.

[4] Braganza DM, Bennett MR. New insights into atherosclerotic plaque rupture.Postgrad Med J 2001;77:94e8.

[5] Valdivielso JM, Coll B, Fernandez E. Vitamin D and the vasculature: can weteach an old drug new tricks? Expert Opin Ther Targets 2009;13:29e38.

[6] Valdivielso JM, Cannata-Andia J, Coll B, Fernandez E. A new role for vitamin Dreceptor activation in chronic kidney disease. Am J Physiol Renal Physiol2009;297:F1502e9.

[7] Li YC, Kong J, Wei MJ, Chen ZF, Liu SQ, Cao LP. 1,25-Dihydroxyvitamin D-3 is anegative endocrine regulator of the renin-angiotensin system. J Clin Invest2002;110:229e38.

[8] Yuan W, Pan W, Kong J, Zheng W, Szeto FL, Wong KE, et al. 1,25-dihydroxyvitamin D3 suppresses renin gene transcription by blocking theactivity of the cyclic AMP response element in the renin gene promoter. J BiolChem 2007;282:29821e30.

[9] Maccarthy EP, Yamashita W, Hsu A, Ooi BS. 1,25-Dihydroxyvitamin-D3 and ratvascular smooth-muscle cell-growth. Hypertension 1989;13:954e9.

[10] Song YF, Qi HW, Wu CG. Effect of 1,25-(OH)(2)D-3 (a vitamin D analogue) onpassively sensitized human airway smooth muscle cells. Respirology 2007;12:486e94.

[11] Chen SC, Law CS, Grigsby CL, Olsen K, Gardner DG. A role for the cell cyclephosphatase Cdc25a in vitamin D-dependent inhibition of adult rat vascularsmooth muscle cell proliferation. J Steroid Biochem Mol Biol 2010;122:326e32.

[12] Mitsuhashi T, Morris RC, Ives HE. 1,25-Dihydroxyvitamin-D3 Modulatesgrowth of vascular smooth-muscle cells. J Clin Invest 1991;87:1889e95.

[13] Koh E, Morimoto S, Fukuo K, Itoh K, Hironaka T, Shiraishi T, et al. 1,25-Dihydroxyvitamin-D3 binds specifically to rat vascular smooth-muscle cellsand stimulates their proliferation in vitro. Life Sci 1988;42:215e23.

[14] Mohtai M, Yamamoto T. Smooth muscle cell proliferation in the rat coronaryartery induced by vitamin D. Atherosclerosis 1987;63:193e202.

[15] Cardus A, Parisi E, Gallego C, Aldea M, Fernandez E, Valdivielso JM. 1,25-dihydroxyvitamin D-3 stimulates vascular smooth muscle cell proliferationthrough a VEGF-mediated pathway. Kidney Int 2006;69:1377e84.

P. Valcheva et al. / Atherosclerosis 235 (2014) 247e255 255

[16] Cardus A, Panizo S, Encinas M, Dolcet X, Gallego C, Aldea M, et al. 1,25-Dihydroxyvitamin D(3) regulates VEGF production through a vitamin Dresponse element in the VEGF promoter. Atherosclerosis 2009;204:85e9.

[17] Cardus A, Panizo S, Parisi E, Fernandez E, Valdivielso JM. Differential effects ofvitamin D analogues on vascular calcification. J Bone Miner Res 2007;22:860e6.

[18] Panizo S, Cardus A, Encinas M, Parisi E, Valcheva P, Lopez-Ongil S, et al. RANKLincreases vascular smooth muscle cell calcification through a RANK-BMP4-dependent pathway. Circ Res 2009;104:1041e8.

[19] Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, et al. Micelacking the vitamin D receptor exhibit impaired bone formation, uterine hy-poplasia and growth retardation after weaning. Nat Genet 1997;16:391e6.

[20] Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, et al.Characterization of a vitamin D receptor knockout mouse as a model ofcolorectal hyperproliferation and DNA damage. Carcinogenesis 2001;22:1429e35.

[21] Parisi E, Almaden Y, Ibarz M, Panizo S, Cardus A, Rodriguez M, et al. N-methyl-D-aspartate receptors are expressed in rat parathyroid gland and regulate PTHsecretion. Am J Physiol Renal Physiol 2009;296:F1291e6.

[22] Dimri GP, Lee XH, Basile G, Acosta M, Scott C, Roskelley C, et al. A biomarkerthat identifies senescent human-cells in culture and in aging skin in-vivo. ProcNatl Acad Sci U S A 1995;92:9363e7.

[23] Sanchez-Nino MD, Bozic M, Cordoba-Lanus E, Valcheva P, Gracia O, Ibarz M,et al. Beyond proteinuria: VDR activation reduces renal inflammation inexperimental diabetic nephropathy. Am J Physiol Renal Physiol 2012;302:F647e57.

[24] Min LJ, Mogi M, Iwanami J, Li JM, Sakata A, Fujita T, et al. Cross-talk betweenaldosterone and angiotensin II in vascular smooth muscle cell senescence.Cardiovasc Res 2007;76:506e16.

[25] Covic A, Mardare N, Gusbeth-Tatomir P, Brumaru O, Gavrilovici C,Munteanu M, et al. Increased arterial stiffness in children on haemodialysis.Nephrol Dial Transplant 2006;21:729e35.

[26] Forster RE, Jurutka PW, Hsieh JC, Haussler CA, Lowmiller CL, Kaneko I, et al.Vitamin D receptor controls expression of the anti-aging klotho gene inmouse and human renal cells. Biochem Biophys Res Commun 2011;414:557e62.

[27] Eelen G, Verlinden L, Meyer MB, Gijsbers R, Pike JW, Bouillon R, et al. 1,25-Dihydroxyvitamin D(3) and the aging-related forkhead box O and sestrinproteins in osteoblasts. J Steroid Biochem Mol Biol 2013;136:112e9.

[28] Kim MY, Kang ES, Ham SA, Hwang JS, Sik T, Lee H, et al. The PPAR delta-mediated inhibition of angiotensin II-induced premature senescence in hu-man endothelial cells is SIRT1-dependent. Biochem Pharmacol 2012;84:1627e34.

[29] Dunlop TW, Vaisanen S, Frank C, Molnar F, Sinkkonen L, Carlberg C. The hu-man peroxisome proliferator-activated receptor delta gene is a primary targetof 1 alpha,25-dihydroxyvitamin D-3 and its nuclear receptor. J Mol Biol2005;349:248e60.

[30] Fukuda N, Satoh C, Hu WY, Soma M, Kubo A, Kishioka H, et al. Production ofangiotensin II by homogeneous cultures of vascular smooth muscle cells fromspontaneously hypertensive rats. Arterioscler Thromb Vasc Biol 1999;19:1210e7.

[31] Hackenthal E, Hackenthal R, Hilgenfeldt U. Isorenin, pseudorenin, cathepsin Dand renin. A comparative enzymatic study of angiotensin-forming enzymes.Biochim Biophys Acta 1978;522:574e88.

[32] Hetman M, Perschl A, Saftig P, Von Figura K, Peters C. Mouse cathepsin Dgene: molecular organization, characterization of the promoter, and chro-mosomal localization. DNA Cell Biol 1994;13:419e27.

[33] UshioFukai M, Zafari AM, Fukui T, Ishizaka N, Griendling KK. p22(phox) is acritical component of the superoxide-generating NADH/NADPH oxidase sys-tem and regulates angiotensin II-induced hypertrophy in vascular smoothmuscle cells. J Biol Chem 1996;271:23317e21.

[34] Mitra AK, Gao L, Zucker IH. Angiotensin II-induced upregulation of AT(1) re-ceptor expression: sequential activation of NF-kappa B and Elk-1 in neurons.Am J Physiol Cell Physiol 2010;299:C561e9.

[35] Kunieda T, Minamino T, Nishi JI, Tateno K, Oyama T, Katsuno T, et al. Angio-tensin II induces premature senescence of vascular smooth muscle cells andaccelerates the development of atherosclerosis via a p21-dependent pathway.Circulation 2006;114:953e60.

[36] Touyz RM, Schiffrin EL. Signal transduction mechanisms mediating thephysiological and pathophysiological actions of angiotensin II in vascularsmooth muscle cells. Pharmacol Rev 2000;52:639e72.

[37] Tsugu A, Sakai K, Dirks PB, Jung S, Weksberg R, Fei YL, et al. Expression ofp57(KIP2) potently blocks the growth of human astrocytomas and inducescell senescence. Am J Pathol 2000;157:919e32.

[38] Gordon D, Reidy MA, Benditt EP, Schwartz SM. Cell proliferation in humancoronary arteries. Proc Natl Acad Sci U S A 1990;87:4600e4.

[39] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s.Nature 1993;362:801e9.

[40] Minamino T, Yoshida T, Tateno K, Miyauchi H, Zou YZ, Toko H, et al. Ras in-duces vascular smooth muscle cell senescence and inflammation in humanatherosclerosis. Circulation 2003;108:2264e9.

[41] Wang JC, Bennett M. Aging and atherosclerosis mechanisms, functional con-sequences, and potential therapeutics for cellular senescence. Circ Res2012;111:245e59.

[42] Coll B, Betriu A, Martinez-Alonso M, Borras M, Craver L, Amoedo ML, et al.Cardiovascular risk factors underestimate atherosclerotic burden in chronickidney disease: usefulness of non-invasive tests in cardiovascular assessment.Nephrol Dial Transplant 2010;25:3017e25.

[43] Coll B, Betriu A, Martinez-Alonso M, Amoedo ML, Arcidiacono MV, Borras M,et al. Large artery calcification on dialysis patients is located in the intima andrelated to atherosclerosis. Clin J Am Soc Nephrol 2011;6:303e10.

[44] Szeto FL, Reardon CA, Yoon D, Wang YL, Wong KE, Chen YZ, et al. Vitamin Dreceptor signaling inhibits atherosclerosis in mice. Mol Endocrinol 2012;26:1091e101.