Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

Contents lists available at ScienceDirect

Journal of Molecular and Cellular Cardiology journal homepage: www.elsevier.com/locate/yjmcc

Original article

Carbonic anhydrase XII in valve interstitial cells promotes the regression of calcific aortic valve stenosis Rihab Bouchareb a, Nancy Côté a, Marie-Chloé-Boulanger a, Khai Le Quang d, Diala El Husseini a, Jérémie Asselin b, Fayez Hadji a, Dominic Lachance d, Elnur Elyar Shayhidin a, Ablajan Mahmut a, Philippe Pibarot d, Yohan Bossé c, Younes Messaddeq b, Denis Boudreau b, André Marette d, Patrick Mathieu a,⁎ a Laboratoire d'Études Moléculaires des Valvulopathies (LEMV), Groupe de Recherche en Valvulopathies (GRV), Quebec Heart and Lung Institute/Research Center, Department of Surgery, Laval University, Québec, Canada b The Center for Optics, Photonics and Lasers (COPL), Department of Physics, Laval University, Québec, Canada c Department of Molecular Medicine, Laval University, Québec, Canada d Department of Medicine, Laval University, Québec, Canada

a r t i c l e

i n f o

Article history: Received 12 December 2014 Received in revised form 20 February 2015 Accepted 2 March 2015 Available online 11 March 2015 Keywords: Calcific aortic valve disease Calcific aortic stenosis Carbonic anhydrase XII P2Y2 receptor Mineral resorption Mineral regression

a b s t r a c t Aims: Calcific aortic valve stenosis (CAVS) is the most common heart valve disease. In the present work we sought to determine the reversibility of mineralization in the aortic valve. Methods and results: By using in vitro analyses we found that valve interstitial cells (VICs) have the ability to resorb minerals. We documented that agonist of P2Y2 receptor (P2Y2R) promoted the expression of carbonic anhydrase XII (CAXII) at the cell membrane of VICs, whereby minerals are resorbed. P2Y2R-mediated mineral resorption was corroborated by using mouse VICs isolated from wild type and P2Y2R−/− mice. Measurements of extracellular pH (pHe) by using core–shell nanosensors revealed that P2Y2R-mediated CAXII export to the cell membrane led to an acidification of extracellular space, whereby minerals are resorbed. In vivo, we next treated LDLR−/−/ApoB100/100/IGF2 mice, which had developed CAVS under a high-fat/high-sucrose diet for 8 months, with 2-thioUTP (a P2Y2R agonist) or saline for the next 2 months. The administration of 2-thioUTP (2 mg/kg/day i.p.) reduced the mineral volume in the aortic valve measured with serial microCT analyses, which improved hemodynamics and reduced left ventricular hypertrophy (LVH). Examination of leaflets at necropsy confirmed a lower level of mineralization and fibrosis along with higher levels of CAXII in mice under 2-thioUTP. In another series of experiment, the administration of acetazolamide (a CA inhibitor) prevented the acidification of leaflets and the regression of CAVS induced by 2-thioUTP in LDLR−/−/ApoB100/100/IGF2 mice. Conclusion: P2Y2R-mediated expression of CAXII by VICs acidifies the extracellular space and promotes the regression of CAVS. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Calcific aortic valve stenosis (CAVS) is a chronic disorder characterized by an abnormal mineralization of aortic leaflets [1]. Although CAVS can be diagnosed in its early stage with echocardiographic examination, there is no medical treatment that can prevent the progressive calcification of leaflets, which ultimately results in a severe stenosis [2]. Hence, the only treatment of CAVS is thus a surgical intervention, which consists in the replacement of the aortic valve by using either an open heart surgery or a percutaneous intervention. However,

⁎ Corresponding author at: Institut de Cardiologie et de Pneumologie de Québec/ Quebec Heart and Lung Institute, 2725 Chemin Ste-Foy Québec, Québec G1V 4G5, Canada. Tel.: +1 418 656 4717; fax: +1 418 656 4707. E-mail address: [email protected] (P. Mathieu).

http://dx.doi.org/10.1016/j.yjmcc.2015.03.002 0022-2828/© 2015 Elsevier Ltd. All rights reserved.

these interventions are invasive and are associated with a significant morbidity/mortality as well as with elevated cost. Bone mineralization is the result of a delicate balance between deposition and resorption of minerals [3]. If applied to pathologic mineralization this concept suggests that it could be possible to promote the resorption of ectopic calcium deposit. Emerging evidence suggests that the expression of carbonic anhydrase (CA) during vascular calcification may promote the resorption of minerals [4]. In the reversa LDLR−/− ApoB100/100 mice the normalization of cholesterol level after 6 months of western type diet led to a significant reduction of mineral content in the aortic valve [5]. These data suggest that ectopic mineralization of the aortic valve is potentially reversible. However, the molecular mechanisms that may promote the regression of aortic valve mineralization are presently unknown. The purinergic system exerts an important control over pathologic mineralization of the aortic valve [6]. In CAVS, the overexpression of ectonucleotidase enzymes,

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

which use nucleotides as substrates, contribute to deplete the extracellular level of nucleotides and to decrease purinergic signaling delivered to valve interstitial cells (VICs), the main cellular component of the aortic valve. P2Y2 receptor (P2Y2R) inhibits the osteogenic transition of VICs and the blockade of ectonucleotidases prevents the development of CAVS [7,8]. However, whether a purinergic receptor-mediated process could promote the regression of CAVS remains to be explored. Herein, we report a novel function for VICs, which have the intrinsic capacity to resorb minerals. We show through several lines of evidence that stimulation of P2Y2R promotes the transport of carbonic anhydrase XII to the cell membrane, whereby pathologic mineralization of the aortic valve is resorbed.

105

The semi-quantitative morphometric analysis of pits formation was evaluated by Von Kossa staining of cultures or by scanning electron microscopy. 2.6. Scanning electron microscopy Samples were fixed 2 h in 2.5% glutaraldehyde at 4 °C and washed in 0.1 M cacodylate before being post-fixed with osmium tetroxide 1% for 1 h at 4 °C. Dehydration was then performed with increasing ethanol concentrations up to the critical point of drying with hexamethyldisilazane overnight. Dried samples were sputtered with palladium (Nanotech, USA) and observed by scanning electron microscopy (SEM) at 30 kV (Quanta FEG 3D, FEI, USA).

2. Material and methods 2.7. Real-time PCR Expanded material and method sections are in the online supplementary material. 2.1. Reagents Dulbecco's modified Eagle's medium (DMEM) and fetal bovine serum (FBS) were purchased from Invitrogen Life Technologies/Thermo Fisher Scientific (ON, Canada). Concentrations used were: ARL67156 (25–100 μM), (ectonucleotidases inhibitor/Tocris Bioscience, MI, USA), 2-thioUTP (2–20 μM regression) (P2Y2 receptor agonist/ Tocris Bioscience, MI, USA or Tri-Link BioTechnologies, CA, USA), ATPγS (50–100 μM) (non-hydrolyzable ATP/Tocris Bioscience, MI, USA), MRS2365 (2 nM) (selective P2Y1 receptor agonist/Tocris Bioscience, MI, USA), acetazolamide (20–40 μM) (carbonic anhydrase inhibitor/SigmaAldrich, ON, Canada), SITS (50 μM) (4-Acetamido-4′-isothiocyanato2,2′-stilbenedisulfonic acid disodium salt hydrate) (anion transport inhibitor/Sigma-Aldrich, ON, Canada). 2.2. Antibodies Antibodies against CAXII (D-2) (monoclonal, sc-374314), CAXII (H-114) (polyclonal, sc-25601), CAII (G-2) (sc-48351) and calnexin (H-70) (sc-11397) were obtained from Santa Cruz Biotechnologies (TX, USA). Antibodies against α-actin (A2547), β-actin (A2228) and vimentin (V2258) were from Sigma-Aldrich (ON, Canada). Antibodies against von Willebrand factor (factor VIII) (250642) were from Abbiotec (CA, USA).

Quantitative real-time PCR (q-PCR) was performed with QuantiTect SYBR Green PCR kit from Qiagen on the Rotor-Gene 6000 system (Corbett Robotics Inc, CA, USA). 2.8. Immunofluorescence of cells Slides were mounted and analyzed with an Ultraview spinning disk confocal imaging system (objective 100× oil, 1.4 NA, PerkinElmer Life and Analytical Sciences, MA, USA) equipped with a cooled electron multiplying charge-coupled device camera at − 50 °C (Hamamatsu Photonics K.K., Hamamatsu-shi, Japan) and driven by Volocity software, version 6.0.1 (PerkinElmer Life and Analytical Sciences). 2.9. Nanogold immunolabeling Mouse VICs were seeded on glass coverslips. Before fixation, cells were left untreated or treated for 24 h with 2-thioUTP. Incubation with anti-CAXII (Santa Cruz Biotechnologies, TX, USA) was performed in the same solution overnight at 4 °C. Dried samples were sputtered with carbon (Nanotech, USA) and observed by scanning electron microscopy (SEM) at 30 kV (Quanta FEG 3D, FEI, USA). 2.10. Measurement of extracellular pH The extracellular pH was measured by fluorescence ratio imaging of fluorescein-conjugated to core–shell nanobiosensors.

2.3. Valve interstitial cell isolation and in vitro analyses of calcification

2.11. Measurement of cytosolic pH by flow cytometry

Control non-calcified aortic valves, used for cell culture, were obtained during heart transplant procedure. Consents were obtained from patients; this protocol (2012–1984, 20770) is approved by the local ethics committee and was performed according to the declaration of Helsinki. Human and mouse (WT and P2Y2R−/−, approved by Laval University animal ethics committee, protocol 2012173) (The Jackson Laboratory, USA) valve interstitial cells (VICs) were isolated by collagenase type I digestion (Invitrogen Life Technologies/Thermo Fisher Scientific, ON, Canada).

Cells were incubated with 2 μM BCECF-AM (Molecular Probes/ Thermo Fisher Scientific, ON, Canada) in DMEM medium at 37 °C for 30 min in the presence of 5% CO2 in dark. Cells were then transferred into flow cytometry tubes for the measurement of fluorescence. FL1/FL2 ratios were calculated using the fluorescence mean values to create standard curve. The ratio was used to create the linear equation obtained from the standard curve to calculate pHi.

2.4. Determination of calcium concentrations

All animal protocols were conducted according to guidelines set out by the Laval University Animal Care and Handling Committee (protocol 2012175) and are conform to the NIH guidelines for the care and use of laboratory animals. LDLR−/− /ApoB100/100 /IGF2 (on C57Bl/6J background) were generated from an established colony at the Heart and Lung Institute of Laval University from original founders kindly provided by Dr. SeppoYlä-Herttuala (University of Eastern Finland, Finland). Male mice were housed in a pathogen-free, temperaturecontrolled environment under a 12:12 hour light–dark cycle and fed ad libitum of a high fat, high sucrose, cholesterol diet (55% calories from fat, 28% from sucrose, 0.2% cholesterol) for 10 months starting at

Calcium content was determined by the Arsenazo III method (Synermed, CA, USA), which relies on the specific reaction of Arsenazo III with calcium to produce a blue complex. 2.5. Mineral resorption studies in osteologic discs Cells were seeded on osteologic discs (BD Biosciences, ON, Canada) or on OsteoAssay Surface (Corning, NY, USA) culture plates. VICs were cultured into 24-well plates containing osteologic discs for 21 days.

2.12. Animals

106

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

12 weeks of age. At 8 months, mice with CAVS were allocated different treatments. A CAVS was considered significant when the peak transaortic velocity was ≥ 160 cm/s. At the end of protocol, mice

were sacrificed by anesthesia under isoflurane (2–3%, inhalation) and cardiac puncture, which was performed by a qualified animal care technician.

Fig. 1. VICs have a potential to regress minerals. a) Schemes depicting the experimental design of the regression assay. VICs were treated with the mineralizing medium for 7 days and then switched to an intervention for another 7 days. Calcium were measured at days 7 (Ca7) and 14 (Ca14). A regression of mineralization was documented if there was a significant decrease of calcium between days 7 (Ca7) and 14 (Ca14). b) Graph showing the induction of regression in human VIC culture following a treatment with ARL67156 (ARL), ATPγS, and 2-thioUTP (2tUTP) (a P2Y2R agonist), but not with MRS2365 (a P2Y1R agonist) (n = 6). c) 2-thioUTP and ARL67156 induce regression of mineralization in P2Y2R+/+ but not in P2Y2R−/− mouse VICs (MVICs) (n = 5). d and e) The effect of ARL67156, ATPγS and 2-thioUTP on the induction of resorption pits (white areas) in osteologic discs by HVICs (treated for 21 days) (n = 4). Scale bar 50 μm. f) Scanning electron microscopy images of control (CTL) and 2-thioUTP treated VICs on osteologic discs; control VICs did not demineralize osteologic discs as shown by the presence of calcium and phosphorus on energy dispersive X-ray analyses (EDX) (red arrow indicate where the measurement was taken), whereas 2-thioUTP-treated cells produced grooves on osteologic discs (red arrow), which were decalcified as demonstrated by EDX analyses of these regions (n = 4). Scale bars 10 μm. g–i) Focused ion beam technique (FIB) was used to create a trench of 14 μm in demineralized area of a 2-thioUTP-treated cell; then by using scanning electron microscopy (SEM) in the area of interest (inset) we documented that mineralized microaggregates were adherent to the cell membrane. Scale bars, left image 20 μm, right image 1 μm. h) EDX analyses confirmed that the mineralized material (red arrow in panel g) was present on the cell membrane (m); i) represents EDX analyses in the cytosol (c) (white arrow in panel g). Data are mean ± SEM; *p b 0.05 compared to control (CTL); #p b 0.05 compared to Ca7; calculated using analyses of variance (ANOVA).

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

2.13. Echocardiography The investigator performing echocardiography analyses was blinded to the group allocation. Transthoracic echocardiography was performed under 2.5%-isoflurane anesthesia, with the L15-7io (5–12 MHz) and S12-4 (4–12 MHz) probes connected to a Philips HD11XE ultrasound system (Philips Healthcare Ultrasound, Netherlands). 2.14. MicroCT SCAN Calcium depots in the heart and aorta were quantified using the Agatston method with the use of dedicated software (eXplore Microview v.2.2). Only aortic valve calcium was transformed in 3D structure and on a 2D image. Quantifications were performed blinded to the group allocation. 2.15. DEXA Body composition was assessed in all mice using DEXA scan (PIXImus Densitometer; Lunar, GE Medical System). 2.16. Cytokine measurements Cytokine measurements were performed with Milliplex kits (Millipore, MA, USA) on sera extracts using Luminex technology. 2.17. Statistical analyses For the comparisons of groups the results were expressed as means ± SEM. For continuous data, values were compared between groups with Student t-test or ANOVA when two or more than two groups were compared, respectively. Post hoc Tukey analyses were done when the p value of the ANOVA was b0.05. The normality assumption and distributions were tested when appropriate. For mineralization assay statistics were performed on raw data and illustrated in graphs with calcium measured at day 7 (Ca7) as the referent set at 100%. The number of experiments performed in VICs refers to different experiments performed from different donors. A p value b 0.05 was considered as statistically significant. Statistical analysis was performed with a commercially available software package JMP IN 8.1. 3. Results 3.1. VICs resorb minerals through P2Y2 receptor To document the potential of VICs to resorb minerals we first established a cell model in which human valve interstitial cells (HVICs) isolated from non-mineralized leaflets (phenotype presented in Supplementary Fig. 1) were cultured for 7 days with a mineralizing medium containing phosphate and then switched to a treatment acting on purinergic receptors. Measurement of calcium was performed at days 7 (Ca7) and 14 (Ca14) to determine whether or not mineral regression had occurred in cell cultures between days 7 and 14; a lower level of minerals at day 14 (Ca14) when compared to day 7 (Ca7) indicated a regression in mineral content (Fig. 1a). At day 7 the introduction of ARL67156, an ectonucleotidase inhibitor which increases the extracellular level of ATP, promoted a regression in calcium content of cell cultures (Fig. 1b). Among the metabotropic receptors only P2Y1R and P2Y2R are expressed by VICs [6]. The addition at day 7 of ATPγS, a non-hydrolyzable nucleotide, or 2-thioUTP, a specific P2Y2R agonist, led to a regression of calcium content whereas MRS2365, a P2Y1R agonist, did not modify the level of calcium measured at day 14 (Fig. 1b). Experiments were also carried out while maintaining the mineralizing-medium for the whole duration of cell culture during 14 days. In these experiments, 2-thioUTP promoted the regression of calcium content even when the mineralizing medium was maintained

107

during 14 days (Supplementary Fig. 2). We next used wild type mouse VICs (MVICs) and P2Y2R−/− cells in order to corroborate the role of P2Y2R as a promoter of mineral regression (MVICs phenotype is presented in Supplementary Fig. 3). When compared to P2Y2R+/+ cells the level of calcium at day 7 in P2Y2R−/− MVICs was significantly increased (Fig. 1c). In addition, when added at day 7 to the growth medium ATPγS and 2-thioUTP promoted the regression of calcium content in P2Y2R+/+ MVICs, whereas they did not promote the regression of calcium content in P2Y2R−/− MVICs (Fig. 1c). Taken together, these data suggested that P2Y2R promoted the regression of calcium content in a cell culture model. To further substantiate these results we next plated HVICs on osteologic discs covered with calcium and phosphate and determined if regression of minerals could be induced. After 21 days, we found that ARL67156, ATPγS and 2-thioUTP induced the formation of resorption pits (Figs. 1d and e). Scanning electron microscope combined with energy dispersive X-ray (EDX) analyses of osteologic discs treated with 2-thioUTP revealed that demineralized areas were present in the vicinity of HVICs (Fig. 1f). By using dual beam microscopy [focused ion beam microscope (FIB) and SEM] a demineralized region of interest at the proximity of a 2-thioUTP treated cell was dissected by creating linear openings distanced from 14 μm and perpendicular to the cell membrane (Fig. 1g). After dissecting the region of interest we observed by using the SEM-EDX that mineralized microaggregates were closely associated with the cell membrane suggesting that demineralization was active at the cell surface of VICs (Figs. 1g–i). Hence, by using different approaches we documented that VICs have the ability to resorb minerals in a P2Y2R dependent manner and that the resorption process likely takes place at the interface of cell membrane and minerals. 3.2. VIC-induced resorption of minerals is dependent on carbonic anhydrase XII We next evaluated the expression of osteoclast markers during a treatment with 2-thioUTP. In basal condition, VICs expressed vimentin and α-smooth muscle actin, which is typical for this cell population (Supplementary Fig. 1) [9]. In both basal and under 2-thioUTP conditions VICs neither expressed tartrate acid resistant phosphatase (TRAP) (Supplementary Figs. 1b and c), osteoclast-associated immunoglobulin-like receptor (OSCAR), CAII nor the receptor activator of nuclear factor kappa-B ligand (RANKL) (Supplementary Figs. 1a and d). However, osteopontin was expressed by VICs but its level was not modified by 2-thioUTP (Supplementary Fig. 1d). Hence, the mineral resorption induced by VICs is not related to a formal switch to an osteoclast phenotype. We next hypothesized that the resorption of minerals induced by VICs is likely the result of an acidification process, which may be induced by a carbonic anhydrase (CA). Treatment of HVICs with acetazolamide, an inhibitor of CAs, negated the regression of mineral content induced by 2-thioUTP in HVIC cultures (Fig. 2a). Similarly, in the osteologic disc model acetazolamide prevented the 2-thioUTPinduced formation of resorption pits (Fig. 2b). These experiments suggested that P2Y2R-induced resorption of minerals in VICs relied on CA. We next performed a gene profiling of all CAs (CAI–CAXV) and documented that CAVB, CAIX, CAXII and CAXIII were expressed by VICs (Fig. 2c). A silencing of each CA (Supplementary Fig. 4) showed that only a siRNA against CAXII, a membrane-bound enzyme, increased mineralization when exposed to a phosphate-containing medium (Fig. 2d). Also, a transfection of CAXII in HVICs reduced by 95% the level of calcium following a treatment with the mineralizing medium (Fig. 2e). These data suggested that CAXII prevents the mineralization of VIC cultures and may thus promote the resorption of minerals. To test this hypothesis, we next silenced CAXII and documented the effect of 2-thioUTP on the regression of calcium content in cell culture. A siRNA targeting CAXII negated the regression of calcium content induced by a P2Y2R agonist, 2-thioUTP (Fig. 2f). Also, the transfection of CAXII in VICs at day 7 promoted a regression of mineralization (Fig. 2g). Confocal scanning

108

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

109

Fig. 3. Activation of CAXII promotes extracellular acidification and regression of minerals. a) Decrease of extracellular pH (pHe) in response to 2-thioUTP (2tUTP) treatment in P2Y2R+/+ MVICs. In P2Y2R−/− MVICs, 2-thioUTP (2tUTP) had no effect on pHe (measurements taken after 24 h of treatment) (n = 4). b) The 2-thioUTP-induced extracellular acidification was prevented by acetazolamide (ACTZ) and by a siRNA for CAXII (n = 4). c) Treatment with 2-thioUTP increases intracellular pH (measurements taken after an equilibration period of 10 min) (n = 5). d) The 2-thioUTP-mediated extracellular acidification was abolished by SITS, a blocker of bicarbonate transporters (n = 5). e) The regression of minerals induced by 2-thioUTP was abolished by SITS (n = 4). Data are mean ± SEM; *p b 0.05 compared to control (CTL); #p b 0.05 compared to Ca7; calculated using analyses of variance (ANOVA).

microscope analyses demonstrated a recruitment of CAXII towards the cell membrane following stimulation with 2-thioUTP in mouse VICs (MVICs) P2Y2R+/+ (Fig. 2h). Quantitative analyses showed that 2thioUTP treatment increased significantly the cellular surface occupied by CAXII (Fig. 2i). Conversely, in MVICs P2Y2R−/− a treatment with 2-thioUTP did not alter CAXII expression and localization, which remained mostly localized in the cytosol (Figs. 2h and i). We next confirmed in MVICs P2Y2R+/+ that 2-thioUTP increased membrane bound CAXII by using confocal microscopy and fluorescent wheat germ agglutinin (WGA), a membrane marker (Supplementary Fig. 5). In addition, by using scanning electron microscope (SEM) we documented that a treatment with the P2Y2R agonist, 2-thioUTP, increased the immunogold labeling of CAXII, which appears to form clusters at the plasma membrane in microdomains (Fig. 2j). Next, by using flow cytometry in MVICs we documented that 2-thioUTP increased the cell membrane expression of CAXII by 2.7-fold (Fig. 2k) (Supplementary Fig. 6). We also measured carbonic anhydrase enzyme activity following fractionation of HVICs and found that 2-thioUTP significantly increased CA activity in the membrane fraction of cells (Fig. 2l). Taken together, this

series of experiments indicates that P2Y2R promotes the export of CAXII to the plasma membrane, whereby resorption of minerals is promoted. 3.3. CAXII promotes extracellular acidification and resorption of minerals The resorption of minerals is likely the result of extracellular acidification promoted by CAXII. To measure the extracellular pH (pHe) we used the fluorescence ratio imaging of unquenched fluorescein (FITC)attached to core–shell nanobiosensors, which allow measurement of pHe in a nanospace environment [10]. The ratiometric method with FITC allows the measurement of pHe within a range of 5–8. Of significance, treatment of MVICs P2Y2R+/+ with the P2Y2R agonist, 2-thioUTP, promoted the acidification of extracellular space decreasing pHe value from 7.52 down to 5.58 (range 5.58–6.5) (Fig. 3a). In P2Y2R−/− MVICs, 2-thioUTP did not induce an acidification of the extracellular space (Fig. 3a). Acetazolamide (a CA inhibitor) and a siRNA for CAXII negated the 2-thioUTP-induced extracellular acidification (Fig. 3b). CA catalyzes + the hydration of carbon dioxide: H2O + CO2 ⇄ HCO− 3 + H . Hence, in order to acidify the extracellular space we hypothesized that the action

Fig. 2. Regression of minerals is promoted by CAXII. a) In cell culture, the addition of 2-thioUTP (2tUTP) at day 7 decreased mineralization at day 14 (Ca14) when compared to day 7 (Ca7), indicating that regression of mineralization had occurred; this effect of 2-thioUTP was abolished by acetazolamide (ACTZ), a CA inhibitor (n = 4). b) In osteologic discs 2-thioUTP (2tUTP) induced the formation of resorption pits, which was negated by ACTZ (n = 4). Scale bar 50 μm. c) A gene profiling of CAs in VICs showed that CAVB, CAIX, CAXII and CAXIII are expressed. d) Silencing of CAs showed that only a siRNA for CAXII increased the mineralization of VICs (mineralization assay over 7 days) (n = 3). e) The transfection of a vector encoding for CAXII in VICs prevented the mineralization of VIC cultures induced by phosphate (mineralizing medium) (n = 4). f) Following a treatment of HVICs with a siRNA against CAXII we documented that 2-thioUTP did not induce a regression of mineralization at day 14 (Ca14) (n = 4). g) Transfection of CAXII at day 7 led to a regression of minerals measured at day 14 (n = 3). h) Average intensity stack of confocal immunofluorescence images of CAXII and F-actin in mouse VICs (MVICs). In P2Y2R +/+ cells, 2-thioUTP induced the dispersion of CAXII over a larger area of the cell surface when compared to P2Y2R−/− MVICs. Furthermore, 2-thioUTP promoted localization of CAXII to the cell membrane in P2Y2R+/+ MVICs (see inset) and not in P2Y2R−/− MVICs (cells treated for 10 min with 2-thioUTP). Objective 100× oil, NA 1.4, scale bar 10 μM. i) Graph depicting percentages of cell surface occupied by CAXII in the indicated conditions. j) SEM studies of VICs with immunogold labeling of CAXII (cells treated for 24 h with 2-thioUTP). Scale bar 500 nm. k) Flow cytometer analyses of non-permeabilized VICs confirmed a higher expression of CAXII at the cell membrane following a treatment with 2-thioUTP for 10 min (n = 3). l) Graph showing membrane-specific carbonic anhydrase activity in control and 2-thioUTP treated cells (measurements taken after 48 h of treatment) (n = 3). Data are mean ± SEM; *p b 0.05 compared to control (CTL); #p b 0.05 compared to Ca7; calculated using analyses of variance (ANOVA).

110

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

of CAXII, a membrane-anchored enzyme with an extracellular active site, was coupled to the intracellular channeling of HCO− 3 allowing the accumulation of H+ outside the cell membrane. We next assessed cytosolic intracellular pH (pHi) by using the pH-sensitive fluorescent probe BCECF and dual-emission ratiometric and cytometry analyses [11]. A treatment of P2Y2R+/+ MVICs with 2-thioUTP increased the pHi significantly (Fig. 3c). Also, disulfonic stilbenes (SITS), an anion channel blocker, prevented the 2-thioUTP-induced extracellular acidification in P2Y2R+/+ MVICs (Fig. 3d). A treatment of P2Y2R+/+ MVICs with SITS also prevented the regression of calcium content in cell cultures induced by 2-thioUTP (Fig. 3e). Therefore, these findings suggested that P2Y2Rmediated extracellular acidification and the resorption of minerals are coupled to the intracellular channeling of HCO3−. 3.4. P2Y2 receptor agonist promotes the regression of CAVS in vivo We next tested in vivo if the administration of a P2Y2R agonist (2-thioUTP) could promote the regression of CAVS. LDLR −/− ApoB 100/100 mice overexpressing IGF2 in the pancreatic β cells (LDLR−/−/ApoB100/100/IGF2) and under a high-fat/high-sucrose diet (HF/HS) develop CAVS [12–14]. After 8 months under HF/HS diet the LDLR−/−/ApoB100/100/IGF2 mice had elevated transaortic velocities with a reduced fractional shortening (FS) (Supplementary Table 1). For the next two months mice continued to be given the HF/HS diet and either received 2-thioUTP (2 mg/kg/day i.p.) or saline (i.p.). Mice receiving saline (n = 11) increased their transaortic velocities during the next 2 months, whereas the administration of 2-thioUTP (n = 10) led to a decrease in transaortic velocities (Fig. 4a). In mice treated with 2-thioUTP these modifications were accompanied by a significant improvement in left ventricular function (FS and ejection fraction) (Figs. 4b and c) and with less left ventricular hypertrophy (LVH) (Fig. 4d). Moreover, the myocardial expression of LVH-related genes β-myosin heavy chain (β-MHC), brain natriuretic peptide (BNP) and transforming growth factor β1 (TGF-β1) were lowered in the 2-thioUTP group (Fig. 4e). In 2-thioUTP treated mice, when compared to baseline values these changes were paralleled by a decrease in aortic valve mineral content as measured by microCT (Figs. 4f–h). On the other hand, the aortic valve mineral content increased in mice receiving saline (Figs. 4f–h). The bone mineral density was not affected by the treatment with 2-thioUTP (Fig. 4i). The blood plasma levels of cholesterol, glucose (Supplementary Table 2) and inflammatory cytokines (IL-1α, IL-6 and TNFα) were not modified in mice receiving 2-thioUTP (Fig. 4j). Post-mortem examination indicated that the area occupied by minerals (detected with alizarin red staining and confirmed with EDX) in aortic valves was lower in mice treated with 2-thioUTP compared to the control group (Fig. 5a). Whereas the aortic leaflets of control mice were thickened and contained a high level of collagen (Masson trichrome and picrosirius stainings), the leaflets of 2-thioUTP-treated mice contained less collagen (Figs. 5b and c). The size of atherosclerotic plaque and the media were not affected by 2-thioUTP (Figs. 5d and e). Immunofluorescence studies showed a higher area occupied by CAXII in aortic valves of 2-thioUTP-treated mice compared to control (Figs. 6a and b). SEM-EDX studies showed in aortic leaflets of 2-thioUTP-treated mice that CAXII was expressed at the cell membrane of VICs, which were interacting with spheroid mineralized microparticles (3–5 μm) (Fig. 6c). We next verified if the 2-thioUTP-mediated regression of CAVS relied on CA activity by using the CA inhibitor, acetazolamide (50 mg/kg/day i.p.) [4], which was coadministered with 2-thioUTP. In mice with CAVS (Supplementary Table 3), 2-thioUTP administered for 2 months promoted a regression of transaortic velocities, whereas acetazolamide (n = 6) negated this effect (Fig. 6d). We next measured tissue pH on aortic valve sections by using fluorescent core–shell nanobiosensors, from which pH of tissue section could be determined precisely by the ratiometric method. Fig. 6e shows representative spectral images of pH values for aortic valve sections in each group of mice. Mice treated with 2-thioUTP

showed more acidic areas in aortic valve leaflets when compared to the control group and mice co-treated with 2-thioUTP + acetazolamide. In mice treated with 2-thioUTP, the mean pH of entire leaflet section was acidic (pH 6.4), whereas a co-treatment with acetazolamide negated this effect (Fig. 6f). 4. Discussion In this work, we discovered that agonist of P2Y2R promoted the export of CAXII to the cell membrane of VICs and promoted the acidification of the extracellular space. In turn, CAXII-mediated acidification of the extracellular space promoted the regression of mineralization (Fig. 6g). Hence, the identification of this novel process indicates that biological mechanisms intrinsic to the aortic valve can be activated to reverse ectopic aortic valve mineralization. 4.1. Regression of mineralization in the aortic valve The development of ectopic mineralization can be seen as a balance between pro- and anti-mineralizing factors. While several studies have investigated and emphasized the mechanisms involved in the osteogenic transition of VICs and its contribution to the fibrocalcific remodeling of the aortic valve [15–17], few studies have examined the reversibility of the process. Studies have demonstrated the presence of multinucleated cells, often referred to as osteoclastlike cells, in mineralized atherosclerotic plaques [18]. However, in view of the scarcity of osteoclast-like cells in mineralized vessels/ valves their role as negative regulator of ectopic mineralization remains to be fully investigated. In the present work, we showed that VICs, the main cellular component of valvular tissue, have an ability to reverse mineralization. VICs treated with a P2Y2R agonist maintained a morphology of fibroblast-like cells and did not exhibit markers of osteoclast. Resorption of minerals is coupled to the expression of CA and the acidification of the extracellular space. Mineralization of glutaraldehyde-fixed bovine pericardium implanted subcutaneously is increased in mice deficient for CAII [19]. In a rat model, the activation of CAIV in vascular smooth cells promoted the regression of vascular mineralization [4]. Miller et al. previously reported in the reversa LDLR−/− ApoB100/100 mice that normalization of cholesterol levels after 6 months of western type diet promoted over the next 6 months a reduction of the mineral content in the aortic valve (measured with alizarin red staining at necropsy), but did not impact on the level of fibrosis measured in leaflets [5]. Following the normalization of blood plasma lipid levels in the reversa LDLR−/− ApoB100/100 mice the function of the aortic valve, measured as the cusp separation distance, improved significantly; the LV function and degree of LVH were not reported in the reversa LDLR−/− ApoB100/100 mice. However, the mechanism whereby mineralization regressed in the reversa mice has not been described. In this work, we found that P2Y2R-mediated export of CAXII to the cell membrane of VICs promoted the resorption of minerals. We found in the LDLR−/−/ApoB100/100/IGF2 mice that despite remaining under a HF/HS diet a short term treatment (2 months) with 2-thioUTP, which did not affect blood plasma lipid variables, led to a regression of mineral content of the aortic valve measured on serial microCT. Analyses of aortic valves at necropsy showed a higher level of CAXII, a decreased content of mineral/fibrosis in mice treated with 2-thioUTP when compared to control mice receiving saline. Also, mice treated with 2-thioUTP improved their LV function and had less LVH. In this regard, the explanted hearts of 2-thioUTP treated mice had lower levels of mRNA encoding for β-MHC, BNP and TGFβ1. Moreover, by using nanobiosensors [10], which allowed precise measurement of pH on tissue section, we determined that mice receiving 2-thioUTP acidified their aortic valve. Conversely, the administration of acetazolamide (a CA inhibitor) negated the acidification of aortic leaflets induced by 2-thioUTP treatment. Hence,

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

111

Fig. 4. P2Y2R agonist induces regression of CAVS in a mouse model. a) In LDLR−/−/ApoB100/100/IGF2 mice with CAVS the administration of 2-thioUTP (n = 10) promoted a reduction of transaortic velocity, whereas in the control group (n = 11) treated with saline the transaortic velocity continued to increase. b and c) Mice under 2-thioUTP increased their fractional shortening (FS) (b) and ejection fraction (EF) (c). d) The left ventricular mass indexed to the tibial length was reduced in mice under 2-thioUTP. e) Expression of hypertrophic gene markers was reduced in mice treated with 2-thioUTP. f) The mineral content (volumetric method measured by microCT) increased in control mice and decreased in mice treated with 2-thioUTP. g and h) Representative images of microCT in coronal (g) and axial (h) views of a same mouse before (day 0) and after treatment (day 60). i) The bone mineral density (BMD) of treated mice. j) The blood plasma level of cytokines in control and 2-thioUTP treated mice.

P2Y2R-mediated activation of CAXII promoted extracellular acidification and a reduction in the burden of mineral/fibrotic components of the aortic valve, which improved significantly the hemodynamics of mice. 4.2. Carbonic anhydrase XII In the present study, we documented that the overexpression of CAXII prevented the mineralization of VIC cultures. Moreover, we found that VIC-mediated mineral resorption relied on CAXII. A previous report showed that fibroblast, contrary to a previously held belief, can actively participate to bone mineral resorption [20]. In the latter study, it was documented that fibroblasts can acidify the extracellular space

through vacuolar H+ ATPase (V-ATPase) and an active transport of H+ outside the cell. Hence, along with the present findings these studies support the concept that cells with a mesenchymal lineage such as VICs can acidify the extracellular space and, in doing so, promote the regression of pathologic mineralization. CAXII has been shown to be expressed in some tumor cells where it acidified the extracellular space [21]. It has also been shown to be expressed in the atherosclerotic plaque [22]. It is believed that the expression of membrane associated CAs, such as CAXII, is used by cells as a mechanism to maintain the intracellular pH within a physiological range in hypoxic environment [23]. To this effect, hypoxia-inducible factor 1 alpha (HIF1-α) is a positive regulator of the expression of CAXII [24]. Hence, the overexpression of membrane-associated CAs is associated with the intracellular

112

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

Fig. 5. P2Y2R agonist induces regression of mineralization and fibrosis in mice aortic valve leaflets. a) Evaluation of the amount of minerals in aortic valve leaflets in mice (n = 21) after sacrifice evaluated with alizarin red and SEM-EDX. Scale bar 200 μm in histology panel and scale bar 10 μm in EM study. b and c) The amount of fibrosis in aortic leaflets measured with trichrome Masson (scale bar 200 μm) (b) and picrosirius (scale bar 100 μm) (c) stainings was reduced in mice under 2-thioUTP. d and e) The size of atherosclerotic plaque (d) and of the media (e) in the two groups of mice.

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

113

Fig. 6. Regression of mineralization in vivo relies on acidification of valvular tissues. a and b) Immunofluorescence studies of CAXII in control and 2-thioUTP treated mice. Scale bars; left panel 500 μm, right panel 50 μm. c) SEM of aortic valve sections with immunogold labeling for CAXII showed in 2-thioUTP treated mice the presence of cells interacting with mineralized spheroid microparticles (3–5 μm) (*mineralized particle showed in back scattered electron) where CAXII was expressed (red beads in secondary electron) at the cell membrane (yellow dotted line). Scale bar 4 μm. d) Administration of acetazolamide (ACTZ) (n = 6) abolished the 2-thioUTP-induced regression of transaortic velocities. e) Spectral images of pH in tissue sections of aortic valves. Scale bar 50 μm. f) pH of mice aortic leaflet sections in different groups of mice. g) Scheme depicting the membrane recruitment of CAXII, allowing the acidification of extracellular space whereby minerals are resorbed. *p b 0.05 compared to control (CTL); calculated using analyses of variance (ANOVA).

114

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

channeling of bicarbonate and thus may counteract intracellular acidosis. We found that SITS, a bicarbonate channel inhibitor, prevented the acidification of extracellular space and prevented the regression of mineralization induced by a P2Y2R agonist. These findings were corroborated by showing that stimulation of P2Y2R was associated with a rise of pHi. Taken together, these data indicate that in order to acidify the extracellular space bicarbonate must be channeled within the intracellular compartment. Hence, although still a hypothesis it is possible that CAXII is expressed by VICs as a mean to prevent intracellular acidosis in the relatively hypoxic environment of the aortic valve, which is avascular and relies solely on a diffusion process to extract oxygen.

Author contributions R.B., N.C., M.C.B. and P.M. conceived and designed experiments. P.M. wrote the manuscript. K.LQ. performed echo analyses. P.P. supervised echo and microCT analyses. R.B., M.C.B., N.C., D.L., F.H., and A.M. performed q-PCR, western blot analyses and cell culture experiments. R.B. performed SEM analyses. R.B. performed pH measurements in isolated cells. M.C.B. performed proximity ligation assay, immunofluorescence and immunoprecipitation studies. R.B., J.A., Y.M., and D.B. performed FIB/EDX analyses and measurements of pH on tissue sections. R.B. performed analyses of mice tissues and blood plasma. A. Marette supervised analyses in the mouse model. Y.B. critically reviewed the manuscript. All authors reviewed the manuscript.

4.3. Clinical implications Considering that there is no medical treatment for CAVS [25–27] the present findings may open novel research avenues. In this regard, the build-up of calcium within the aortic valve is regarded has one major culprit in CAVS [28]. In the present series of experiments, we found that an agonist of P2Y2R (2-thioUTP) induced the resorption of aortic valve minerals in mice with CAVS, which resulted in better LV function. Noteworthy, the bone mineral density remained unchanged by the treatment with a P2Y2R agonist, which emphasizes that purinergic response is likely tissue specific. The administration of 2-thioUTP neither modified the inflammatory profile of blood plasma cytokines nor the level of cholesterol. Aortic valves examined at the time of sacrifice had less mineral contents and a lower degree of fibrosis. Taken together the findings of this study provide compelling evidence that VICs have an intrinsic capacity to regress mineralization during CAVS provided appropriate signals are delivered to the cells. Whether only the purinergic receptor or other pathways, which could be manipulated pharmacologically, could also promote the activation of CAXII in VICs remains to be investigated.

4.4. Limitations

Acknowledgments This work was supported by HSFC grant, CIHR grants MOP114893 MOP245048 and Quebec Heart and Lung Institute Fund. N.C. is supported by studentship grants of Fonds de recherche Québec — Santé (FRQS). Y. Bossé is the recipient of a Junior 2 Research Scholar award from the FRQS. P.M. is a senior research scholar from the FRQS, Montreal, Québec, Canada. PP holds the Canada Research Chair in Valvular Heart Disease, Canadian Institutes of Health Research, Ottawa, Ontario, Canada. Disclosures P.M. and N.C. have patent applications for the use of ectonucleotidase inhibitors and P2Y2R agonists in the treatment of CAVD. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yjmcc.2015.03.002. References

−/−

100/100

In the present study LDLR /ApoB /IGF2 mice were treated with 2-thioUTP for a short period of time (2 months) and also relatively early in the disease process (after 8 months of diet). In the reversa LDLR−/− ApoB100/100 mice, the normalization of blood plasma lipids after 12 months although leading to a regression of minerals over the next 6 months was not efficient to improve the function of the aortic valve [29]. Hence, whether treating LDLR−/−/ApoB100/100/IGF2 mice later during the disease process and for longer period of time would lead to important and significant improvement in the hemodynamics is presently unknown and remains to be investigated. Also, the reduction of transaortic velocities was modest and it remains to be determined if a long-term treatment with P2Y2R agonist could lead to a normalization of velocities and a complete regression of mineralization of the aortic valve. Nonetheless, the present study highlighted by using both in vitro and in vivo experiments that P2Y2R-mediated expression of CAXII to the cell membrane of VICs promoted extracellular acidification and the regression of mineralization. Further development in the pharmacology of purinergic receptors [30] and/or identifying novel pathways to enhance membrane expression of CAXII by VICs may offer novel research avenues with potential therapeutic implications.

5. Conclusion We provide compelling evidence that pathologic mineralization of the aortic valve is a reversible process. P2Y2R-mediated activation of CAXII promoted the resorption of minerals. This study thus elucidates a novel mechanism by which mineralization of the aortic valve is controlled and opens novel research perspectives.

[1] Rajamannan NM, Evans FJ, Aikawa E, Grande-Allen KJ, Demer LL, Heistad DD, et al. Calcific aortic valve disease: not simply a degenerative process: a review and agenda for research from the National Heart and Lung and Blood Institute Aortic Stenosis Working Group. Executive summary: calcific aortic valve disease—2011 update. Circulation 2011;124:1783–91. [2] Otto CM. Calcific aortic stenosis—time to look more closely at the valve. N Engl J Med 2008;359:1395–8. [3] Rousselle AV, Heymann D. Osteoclastic acidification pathways during bone resorption. Bone 2002;30:533–40. [4] Essalihi R, Dao HH, Gilbert LA, Bouvet C, Semerjian Y, McKee MD, et al. Regression of medial elastocalcinosis in rat aorta: a new vascular function for carbonic anhydrase. Circulation 2005;112:1628–35. [5] Miller JD, Weiss RM, Serrano KM, Brooks RM, Berry CJ, Zimmerman K, et al. Lowering plasma cholesterol levels halts progression of aortic valve disease in mice. Circulation 2009;119:2693–701. [6] Cote N, El Husseini D, Pepin A, Guauque-Olarte S, Ducharme V, Bouchard-Cannon P, et al. ATP acts as a survival signal and prevents the mineralization of aortic valve. J Mol Cell Cardiol 2012;52:1191–202. [7] El Husseini D, Boulanger MC, Mahmut A, Bouchareb R, Laflamme MH, Fournier D, et al. P2Y2 receptor represses IL-6 expression by valve interstitial cells through Akt: implication for calcific aortic valve disease. J Mol Cell Cardiol 2014;72:146–56. [8] Cote N, El HD, Pepin A, Bouvet C, Gilbert LA, Audet A, et al. Inhibition of ectonucleotidase with ARL67156 prevents the development of calcific aortic valve disease in warfarin-treated rats. Eur J Pharmacol 2012;689:139–46. [9] Mathieu P, Voisine P, Pepin A, Shetty R, Savard N, Dagenais F. Calcification of human valve interstitial cells is dependent on alkaline phosphatase activity. J Heart Valve Dis 2005;14:353–7. [10] Asselin J, Roy C, Boudreau D, Messaddeq Y, Bouchareb R, Mathieu P. Supported core–shell nanobiosensors for quantitative fluorescence imaging of extracellular pH. Chem Commun (Camb) 2014;50:13746–9. [11] Nilsson C, Kagedal K, Johansson U, Ollinger K. Analysis of cytosolic and lysosomal pH in apoptotic cells by flow cytometry. Methods Cell Sci 2003;25:185–94. [12] Heinonen SE, Leppanen P, Kholova I, Lumivuori H, Hakkinen SK, Bosch F, et al. Increased atherosclerotic lesion calcification in a novel mouse model combining insulin resistance, hyperglycemia, and hypercholesterolemia. Circ Res 2007;101: 1058–67. [13] Laplante MA, Charbonneau A, Avramoglu RK, Pelletier P, Fang X, Bachelard H, et al. Distinct metabolic and vascular effects of dietary triglycerides and cholesterol in

R. Bouchareb et al. / Journal of Molecular and Cellular Cardiology 82 (2015) 104–115

[14]

[15]

[16]

[17]

[18] [19] [20]

[21]

atherosclerotic and diabetic mouse models. Am J Physiol Endocrinol Metab 2013; 305:E573–84. Le Quang K, Bouchareb R, Lachance D, Laplante MA, Husseini DE, Boulanger MC, et al. Early development of calcific aortic valve disease and left ventricular hypertrophy in a mouse model of combined dyslipidemia and type 2 diabetes mellitus. Arterioscler Thromb Vasc Biol 2014;34:2283–91. Rajamannan NM, Subramaniam M, Rickard D, Stock SR, Donovan J, Springett M, et al. Human aortic valve calcification is associated with an osteoblast phenotype. Circulation 2003;107:2181–4. Poggio P, Sainger R, Branchetti E, Grau JB, Lai EK, Gorman RC, et al. Noggin attenuates the osteogenic activation of human valve interstitial cells in aortic valve sclerosis. Cardiovasc Res 2013;98:402–10. Chen JH, Chen WL, Sider KL, Yip CY, Simmons CA. Beta-catenin mediates mechanically regulated, transforming growth factor-beta1-induced myofibroblast differentiation of aortic valve interstitial cells. Arterioscler Thromb Vasc Biol 2011; 31:590–7. Doherty TM, Uzui H, Fitzpatrick LA, Tripathi PV, Dunstan CR, Asotra K, et al. Rationale for the role of osteoclast-like cells in arterial calcification. FASEB J 2002;16:577–82. Rajachar RM, Tung E, Truong AQ, Look A, Giachelli CM. Role of carbonic anhydrase II in ectopic calcification. Cardiovasc Pathol 2009;18:77–82. Pap T, Claus A, Ohtsu S, Hummel KM, Schwartz P, Drynda S, et al. Osteoclastindependent bone resorption by fibroblast-like cells. Arthritis Res Ther 2003;5: R163–73. Chiche J, Ilc K, Laferriere J, Trottier E, Dayan F, Mazure NM, et al. Hypoxia-inducible carbonic anhydrase IX and XII promote tumor cell growth by counteracting acidosis through the regulation of the intracellular pH. Cancer Res 2009;69:358–68.

115

[22] Oksala N, Levula M, Pelto-Huikko M, Kytomaki L, Soini JT, Salenius J, et al. Carbonic anhydrases II and XII are up-regulated in osteoclast-like cells in advanced human atherosclerotic plaques—Tampere Vascular Study. Ann Med 2010;42:360–70. [23] Svastova E, Hulikova A, Rafajova M, Zat'ovicova M, Gibadulinova A, Casini A, et al. Hypoxia activates the capacity of tumor-associated carbonic anhydrase IX to acidify extracellular pH. FEBS Lett 2004;577:439–45. [24] Wykoff CC, Beasley NJ, Watson PH, Turner KJ, Pastorek J, Sibtain A, et al. Hypoxia-inducible expression of tumor-associated carbonic anhydrases. Cancer Res 2000;60:7075–83. [25] Cowell SJ, Newby DE, Prescott RJ, Bloomfield P, Reid J, Northridge DB, et al. A randomized trial of intensive lipid-lowering therapy in calcific aortic stenosis. N Engl J Med 2005;352:2389–97. [26] Rossebo AB, Pedersen TR, Boman K, Brudi P, Chambers JB, Egstrup K, et al. Intensive lipid lowering with simvastatin and ezetimibe in aortic stenosis. N Engl J Med 2008; 359:1343–56. [27] Chan KL, Teo K, Dumesnil JG, Ni A, Tam J. Effect of lipid lowering with rosuvastatin on progression of aortic stenosis: results of the aortic stenosis progression observation: measuring effects of rosuvastatin (ASTRONOMER) trial. Circulation 2010;121:306–14. [28] Messika-Zeitoun D, Aubry MC, Detaint D, Bielak LF, Peyser PA, Sheedy PF, et al. Evaluation and clinical implications of aortic valve calcification measured by electron-beam computed tomography. Circulation 2004;110:356–62. [29] Miller JD, Weiss RM, Serrano KM, Castaneda LE, Brooks RM, Zimmerman K, et al. Evidence for active regulation of pro-osteogenic signaling in advanced aortic valve disease. Arterioscler Thromb Vasc Biol 2010;30:2482–6. [30] Mathieu P. Pharmacology of ectonucleotidases: relevance for the treatment of cardiovascular disorders. Eur J Pharmacol 2012;696:1–4.