Abstract
BACKGROUND
Recent clinical evidence suggests an association between warfarin use and calcification of the aortic valve. We sought to determine the effect of warfarin on aortic valve interstitial cell (AVIC) osteogenic protein expression and the signaling pathways by which this effect is mediated.
METHODS
Human AVICs were isolated from normal aortic valves of patients undergoing cardiac transplantation, whereas diseased AVICs were isolated from patients undergoing aortic valve replacement for aortic stenosis. AVICs were treated with various anticoagulants, and osteogenic protein expression was evaluated using immunoblotting. Phosphorylation of lipoprotein receptor-related protein 6 (LRP6) and extracellular signal-regulated kinase 1/2 (ERK1/2) was evaluated after treatment with warfarin. AVICs were pretreated with LRP6 inhibitor dkk1 and ERK1/2 inhibitor PD98059 followed by treatment with warfarin, and osteogenic protein expression was evaluated.
RESULTS
Warfarin, but not heparin or dabigatran, significantly increased Runx-2 and Osx expression in both normal and diseased human AVICs. Upregulation of β-catenin protein expression and nuclear translocation occurred in diseased AVICs but not normal AVICs after warfarin treatment. Warfarin induced phosphorylation of LRP6 in diseased AVICs only and phosphorylation of ERK1/2 in both normal and diseased AVICs. LRP6 inhibition attenuated warfarin-induced Runx-2 expression in diseased AVICs. ERK1/2 inhibition attenuated warfarin-induced Runx-2 expression in both normal and diseased AVICs.
CONCLUSIONS
Warfarin induces osteogenic activity in normal and diseased isolated human AVICs. This effect is mediated by ERK1/2 in both diseased and normal AVICs, but in diseased AVICs β-catenin signaling also plays a role. These results implicate the role of warfarin in aortic valve calcification and highlight potential mechanisms for warfarin-induced aortic stenosis.
Aortic stenosis is a disease of the elderly that carries significant economic and health burdens. A severe form of this disease afflicts over 2% of the population over the age of 75 years, and the prevalence continues to rise.1 Despite the impact of this disease, little is understood concerning the pathogenesis of aortic stenosis. Furthermore the only treatments available for aortic stenosis are invasive procedures, such as surgical or transcatheter techniques to replace the valve. Aortic stenosis has historically been believed to occur through passive degeneration of the valve over time. However recent data now demonstrates that development of aortic stenosis involves active processes of chronic inflammation. A key mediator of the inflammatory pathobiology underlying development of aortic stenosis is the aortic valve interstitial cell (AVIC).2 Prior work has demonstrated that in response to proinflammatory stimuli AVICs can adopt an osteoblast-like phenotype, characterized by production of bone-forming proteins such as runt-related transcription factor-2 (Runx-2) and osterix (Osx).3 These osteogenic phenotypic changes are believed to contribute to the development of aortic stenosis through induction of valvular calcification. This enlightened understanding of the pathobiology behind aortic stenosis enables further investigation into potential therapeutic targets.
Warfarin has been found to induce calcific changes in arteries and heart valves of rodents.4,5 In addition clinical studies have found an association between warfarin use and valvular calcification in patients undergoing anticoagulant therapy.6 These findings are noteworthy because about 3 million people in the United States are currently taking warfarin.7 In a study of vascular smooth muscle cells, Wnt/b-catenin signaling was found to play a critical role in warfarin-induced calcification.8 However the mechanisms underlying warfarin-induced calcification remain elusive. We hypothesized that warfarin can induce osteogenic phenotypic changes in human AVICs and thus may contribute to valvular calcification. We further proposed that Wnt/b-catenin signaling could play an important role. The purpose of this study was to determine the effects of warfarin on osteogenic protein expression in AVICs and the signaling mechanisms through which these effects are mediated.
MATERIAL AND METHODS
This study was approved by the Colorado Multiple Institutional Review Board at the University of Colorado School of Medicine (11–1004).
REAGENTS.
Medium 199, penicillin G, streptomycin, amphotericin B, and Earle’s balanced salt solution were obtained from Lonza (Walkersville, MD). Fetal bovine serum and donkey serum were ordered from Aleken chemicals (Nash, TX). Laemmli sample buffer, nitrocellulose membranes, and 4% to 20% gradient polyacrylamide miniprotean TGX gels were purchased from Bio-Rad (Hercules, CA). Chemiluminescence substrate and Pierce lactate dehydrogenase assay kits were purchased from Thermo Scientific (Rockford, IL). Rabbit-derived anti-human Runx-2 antibody, anti-human β-catenin antibody, anti-human β-actin antibody, anti-phospho/total human low-density-lipoprotein receptor-related protein 6 (LRP6) antibodies, and anti-phospho/total extracellular signal-regulated kinase 1/2 (ERK1/2) were purchased from Cell Signaling (Danvers, MA). Rabbit-derived anti-human Osx antibody was purchased from Santa-Cruz (Dallas, TX). Warfarin was purchased from Sigma Chemicals (St Louis, MO) and consisted of a racemic mixture containing both R- and S-enantiomers. All other reagents and chemicals were purchased from Sigma Chemicals.
CELL ISOLATION AND CULTURE.
Normal-appearing, 3-leaflet aortic valves were procured in the operating room from explanted hearts of patients undergoing cardiac transplantation at the University of Colorado Hospital. Preoperative echocardiography demonstrated no significant aortic valvular disease, and on inspection leaflets were found to be thin, pliable, and without nodules or calcification. Diseased aortic valve tissue was excised from stenotic valves at the time of aortic valve replacement from separate donors. AVICs were isolated as described previously.3 In brief valve tissue was placed in a container filled with sterile saline solution and brought to the laboratory on ice. After washing the leaflets with phosphate-buffered saline (PBS) 5 times, the valve was divided into small pieces using sterile scissors and pickups. The pieces were collected and placed in a 15-mL conical tube along with collagenase (2.5 mg/mL) for 30 minutes. This initial step removed endothelial cells from the surface of the valve leaflet. The supernatant was collected and the tissue digested a second time with collagenase at a concentration of 0.8 mg/mL for 3 hours. Supernatant was removed and centrifuged. Isolated cells from the pellet were then resuspended in cell culture medium (a mixture of M199 with 10% fetal bovine serum, penicillin G, streptomycin, and amphotericin B that was filtered [0.2 μm]) in a small flask (25 cm2). This flask was placed in an incubator maintained at a temperature of 37°C with 5% carbon dioxide. Cells were grown in culture until they reached passages 2 to 6 before use in experiments.
AVIC STIMULATION WITH WARFARIN.
Warfarin was dissolved in sterile PBS with the assistance of heat and gentle agitation. In separate experiments the concentration–response was studied over a range of warfarin concentrations ranging from 5 μM to 20 μM. The maximal response of AVICs to warfarin occurred at a concentration of 10 μM, and so this concentration was chosen for this study. This concentration is also consistent with commonly found therapeutic plasma warfarin levels in vivo, likely corresponding to an international normalized ratio of 2 to 3.9
IMMUNOBLOTTING.
AVICs were plated onto 24-well plates, and after treatments cells were lysed with 1× Laemmli sample buffer with β-mercaptoethanol. Lysates were loaded into 15-well, 4% to 20% gradient miniprotean TGX gels. Gels were run at 180 V for 45 minutes. Protein transfer to nitrocellulose membranes was done using a voltage of 100 V for 70 minutes. An ultraviolet stratalinker (Stratagene, La Jolla, CA) was used to crosslink membranes twice, and then membranes were blocked using 5% dry milk dissolved in 0.1% Tween in PBS (T-PBS) for 45 minutes. After 3 washes with T-PBS the membranes were incubated with primary antibodies of interest at 4°C overnight. The next morning membranes were washed with T-PBS, and then horseradish peroxidase-conjugated secondary antibodies (diluted 1:10,000 in 5% dry milk in T-PBS) were applied for 1 hour. Subsequently membranes were washed 3 times with T-PBS. Membranes were imaged with a Chemi-Doc MP Imaging system (Bio-Rad, Hercules, CA). Densitometry analysis was performed with Image-J software from the National Institutes of Health (Bethesda, MD).
CALCIUM NODULE STAINING.
AVICs were fixed and stained as described previously.10,11 In brief cells were plated onto 12-well plates and treated for 2 weeks in conditioning media (cell culture media as described above in Reagents with the addition of 10 mmol/L β-glycerophosphate, 10 nmol/L vitamin D3, 10 nmol/L dexamethasone, and 8 mmol/L CaCl2). Cells were washed with PBS and fixed in 4% paraformaldehyde for 12 to 16 minutes. The wells were then washed with deionized water 3 times before staining with 0.2% alizarin red solution (pH 4.2) for 30 minutes. After staining, excess dye was rinsed gently with deionized water. Images were taken with a Nikon Eclipse TS100 microscope (Nikon Corporation, Tokyo, Japan) using associated NIS software (version 4.0).
IMMUNOFLUORESCENCE.
AVICs were plated onto 8-chamber slides. Cells were washed with PBS 3 times before being permeabilized with 30% acetone and 70% methanol for 8 minutes. Then cells were washed 2 additional times with PBS before fixing with 4% paraformaldehyde for 8 minutes. The slides were washed with deionized water 3 times and then blocked using donkey serum for 45 minutes. Primary rabbit-derived antibody against β-catenin (diluted 1:80 in 1% bovine serum albumin) was applied to the cells overnight with gentle agitation at 4°C. Slides were then rinsed 3 times with deionized water before incubation with a secondary antibody (Cy3-conjugated donkey anti-rabbit antibody) for 2 hours. Bisbenzimide (4,6-diamidino-2-phenylindole, imaged on the blue channel) nuclear counterstain and wheat germ agglutinin (imaged on the green channel) were also applied over this time period. Chambers were removed, and the slides were washed with deionized water and covered with mounting medium sealed with nail polish. Cells were imaged using a DM5500 B confocal microscope (Leica, Wetzlar, Germany). The β-catenin stain was imaged on the red (Cy3) channel. Application Suite software (Leica) for advanced fluoroscopy was used to obtain images.
STATISTICAL ANALYSIS.
Results are presented as mean ± SEM. Graphpad Prism version 5.0 (La Jolla, CA) was used to perform statistical analysis. The Mann-Whitney U test was used to compare means, and P < .05 was considered a significant difference.
RESULTS
PATIENT INFORMATION.
Normal aortic valve leaflets were obtained from the explanted hearts of 4 patients undergoing transplant. These donors ranged in age from 40 to 61 years, and all were nonsmoking men. The indication for transplantation in all patients was idiopathic dilated cardiomyopathy, and preoperative echocardiogram demonstrated no significant valvular disease. A preoperative angiogram demonstrated no significant coronary artery disease in this group of patients. Diseased, calcified aortic valve leaflets were obtained from 4 patients undergoing aortic valve replacement. The age range for these donors was 54 to 72 years. A preoperative angiogram demonstrated no significant coronary artery disease in these patients. No patients were on warfarin therapy at the time of operation.
WARFARIN-INDUCED OSTEOGENIC ACTIVITY IN DISEASED AND NORMAL HUMAN AVICS.
Warfarin treatment for 48 hours induced increased expression of osteogenic proteins Runx-2 and Osx in both normal and diseased human AVICs (Figure 1A). Densitometry analysis of both Runx-2 and Osx expression revealed that expression of both proteins significantly increased after treatment with warfarin, and this increase was significantly greater in diseased AVICs relative to normal AVICs (Figure 1B).
FIGURE 1.
Warfarin induces an osteogenic phenotype in human aortic valve interstitial cells (AVICs). (A) After stimulation with warfarin for 48 hours the expression of runt-related transcription factor 2 (Runx-2) and osterix (Osx) was determined by immunoblotting in normal and diseased AVICs. (B) Densitometry data demonstrated a significant increase in both Runx-2 and Osx expression (*P < .05). This increase was greater in diseased AVICs relative to normal AVICs (#P < .05) for both Runx-2 and Osx expression.
When compared with normal control human AVICs, treatment with warfarin for 14 days caused an increase in calcium nodule formation (Figure 2). These results demonstrated that warfarin elevated cellular osteogenic activity in human AVICs.
FIGURE 2.
Warfarin induces calcium nodule formation in human aortic valve interstitial cells (AVICs). AVICs were stimulated with conditioning media with or without warfarin for 14 days. Alizarin red calcium nodule staining was performed. Relative to control cells warfarin-treated AVICs demonstrated increased calcium nodule formation.
HEPARIN AND DABIGATRAN DID NOT INDUCE AN OSTEOGENIC PHENOTYPIC CHANGE IN HUMAN AVICS.
Normal and diseased AVICs were treated with heparin (0.1, 1.0, and 10 U/mL) or dabigatran (10, 500, and 1000 ng/mL) for 48 hours. Expression of osteogenic proteins Runx-2 and Osx did not change significantly as determined by immunoblotting (Figure 3A) and corresponding densitometry data (Figure 3B). These results indicated that in contrast to warfarin, heparin and dabigatran did not increase osteogenic activity in either normal or diseased human AVICs.
FIGURE 3.
Heparin and dabigatran do not induce an osteogenic phenotypic change in human aortic valve interstitial cells (AVICs). Normal and diseased AVICs were treated with heparin or dabigatran for 48 hours. (A) Expression of runt-related transcription factor 2 (Runx-2) and osterix (Osx) did not change significantly after heparin or dabigatran stimulation. (B) Corresponding densitometry analysis confirmed these results.
WARFARIN INDUCED AN INCREASE IN β-CATENIN EXPRESSION IN DISEASED AVICS BUT NOT IN NORMAL AVICS.
Warfarin treatment for 48 hours resulted in a significant increase in expression of β-catenin in diseased AVICs but not in normal AVICs (Figure 4). This indicates that diseased or osteogenic AVICs have elevated Wnt/β-catenin signaling activity that contributes mechanistically to warfarin-induced osteogenic phenotypic changes.
FIGURE 4.
Stimulation with warfarin induces an increase in β-catenin expression in diseased aortic valve interstitial cells (AVICs) but not in normal AVICs. Human AVICs were stimulated for 48 hours with warfarin. Immunoblotting and densitometry data suggest that warfarin treatment resulted in a significant increase in β-catenin expression in diseased AVICs but not in normal AVICs.
WARFARIN ACTIVATED β-CATENIN SIGNALING IN DISEASED AVICS BUT NOT IN NORMAL AVICS.
Cells were plated onto chamber slides and treated for 2 hours with warfarin. Slides were then fixed, and immunofluorescent microscopy was performed. Warfarin treatment resulted in nuclear translocation of β-catenin in diseased AVICs but not in normal AVICs (Figure 5).
FIGURE 5.
Warfarin activates β-catenin signaling in diseased aortic valve interstitial cells (AVICs) but not in normal AVICs. Cells were plated onto chamber slides and treated with warfarin for 2 hours followed by immunofluorescent staining. Cell membranes are stained green (wheat germ agglutinin [WGA]), nuclei are stained blue (4′,6-diamidino-2-phenylindole, DAPI), and β-catenin is stained red. In diseased AVICs without stimulation, β-catenin is located at the periphery of the cell (panel C). However with stimulation, β-catenin is translocated into the nucleus (panel G), indicating activation of the β-catenin signaling pathway. In contrast β-catenin is located at the periphery of the normal control AVICs (panel O), but with warfarin stimulation β-catenin remains in the cytosol and along the cell membrane, suggesting the β-catenin pathway is not activated in normal human AVICs.
AVICs were plated and treated with warfarin for 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, and 8 hours. Using immunoblotting, lysates were analyzed for levels of phosphorylated and total LRP6. LRP6 phosphorylation is a key marker of activation of the canonical Wnt/β-catenin signaling pathway. Figure 6 shows that treatment with warfarin significantly increased the fraction of phosphorylated LRP6 in diseased AVICs but not in normal AVICs. This suggests that this portion of the canonical Wnt/β-catenin signaling pathway may be upregulated in diseased versus normal AVICs. Overall these data underscore the mechanistic role for Wnt/β-catenin signaling in warfarin-induced osteogenic activity in diseased AVICs.
FIGURE 6.
Warfarin activates the canonical Wnt/β-catenin pathway signaling in diseased aortic valve interstitial cells (AVICs) but not in normal AVICs. Cells were stimulated with warfarin for various time points, and phosphorylated versus total low-density-lipoprotein receptor-related protein 6 (LRP6) were analyzed. LRP6 phosphorylation is a critical marker of canonical Wnt/β-catenin pathway activation. Warfarin induced phosphorylation of LRP6 in (A) diseased AVICs but not in (B) normal AVICs. These data suggest that canonical Wnt/β-catenin signaling occurs only in diseased cells after treatment with warfarin.
WARFARIN ACTIVATED THE ERK1/2 PATHWAY IN DISEASED AND NORMAL AVICS.
Warfarin treatment for 15 minutes, 30 minutes, 1 hour, 2 hours, and 4 hours resulted in increased levels of phosphorylated ERK1/2 in diseased and normal AVICs (Figure 7). These results show that the ERK1/2 signaling pathway mediates warfarin-induced cellular osteogenic activity in normal and diseased AVICs.
FIGURE 7.
Treatment with warfarin activates the extracellular signal-regulated kinase 1/2 (ERK1/2) signaling pathway in diseased and normal aortic valve interstitial cells (AVICs). Cells were treated with warfarin for various time points. Immunoblotting results show that warfarin induces phosphorylation of ERK1/2 in both (A) diseased and (B) normal AVICs.
ERK1/2 INHIBITION PREVENTS WARFARIN-INDUCED RUNX-2 EXPRESSION IN NORMAL AND DISEASED AVICS, WHEREAS WNT/β-CATENIN INHIBITION PREVENTS THIS RESPONSE IN ONLY DISEASED AVICS.
As shown in Figure 8, warfarin induced a significant increase in Runx-2 expression in diseased and normal AVICs. Inhibition of ERK1/2 (inhibitor PD98059, 5 μM) prevented this increase in Runx-2 expression in both diseased and normal AVICs. However Wnt/β-catenin pathway inhibition (dkk1, 50 ng/mL) prevented warfarin-induced Runx-2 expression in diseased AVICs but not in normal AVICs. These data confirm that the ERK1/2 signaling pathway mediates warfarin-induced cellular osteogenic activity in normal and diseased AVICs, whereas Wnt/b-catenin signaling only played a role in mediating warfarin-induced osteogenic activity in diseased AVICs.
FIGURE 8.
Inhibition of extracellular signal-regulated kinase 1/2 (ERK1/2) attenuates warfarin-induced runt-related transcription factor-2 (Runx-2) expression in diseased and normal aortic valve interstitial cells (AVICs), but Wnt/β-catenin pathway inhibition only attenuates warfarin-induced Runx-2 expression in diseased AVICs. Cells were pretreated with ERK1/2 inhibitor PD98059 and canonical Wnt/β-catenin pathway inhibitor dkk1 before stimulation with warfarin for 48 hours. (A) Both ERK1/2 and Wnt/β-catenin pathway inhibition attenuated the expression of warfarin-induced Runx-2 in diseased AVICs. (B) However only ERK1/2 and not low-density-lipoprotein receptor-related protein 6 inhibition attenuated the increase in Runx-2 expression induced by warfarin in normal AVICs. These data suggest that ERK1/2 signaling is required for induction of Runx-2 by warfarin in both normal and diseased AVICs. In contrast both ERK1/2 and Wnt/β-catenin signaling are necessary for warfarin-induced Runx-2 expression in diseased AVICs.
WARFARIN IS NONTOXIC TO HUMAN AVICS.
A lactate dehydrogenase assay was performed after treating normal AVICs with warfarin for 48 hours at concentrations of 5, 10, and 20 μM (Figure 9). Figure 9 shows that the percentage of toxicity was acceptably low at these warfarin treatment concentrations.
FIGURE 9.
Warfarin treatments are nontoxic to human aortic valve interstitial cells (AVICs). Cells were treated with warfarin at various concentrations for 48 hours. Cell culture media was then analyzed using a lactate dehydrogenase (LDH) assay. Toxicity was acceptably low for warfarin treatments in this concentration range.
COMMENT
Results from this present study demonstrate that warfarin can induce osteogenic phenotypic changes in both diseased and normal AVICs. Warfarin activates ERK1/2 in diseased and normal cells, whereas in diseased cells β-catenin signaling is additionally activated. Both pathways are required for warfarin-induced expression of the critical osteogenic transcription factor Runx-2. These data may have significant implications for clinical practice in the realm of anticoagulant therapy. High-quality clinical studies are needed to confirm in vitro findings and solidify the role of warfarin in aortic valve calcification.
There are several limitations inherent in this study. First this was an in vitro investigation and cannot fully replicate the conditions found in the dynamic in vivo environment of the aortic valve. However we have previously demonstrated that AVICs in passages 2 to 6 display similar behavior to cells isolated directly from the donor.3 We also acknowledge that although the aortic valve leaflets from “normal” donors appear thin and pliable and have no evidence of disease on preoperative echocardiogram, this tissue may contain cells that are not completely phenotypically normal, because these donors suffered from cardiomyopathy. It is important to note our findings pertain only to trileaflet valves, because bicuspid valves may demonstrate underlying pathology that interferes with the effects of warfarin. Aortic valves were obtained from patients who were not taking warfarin such that we could accurately evaluate the effects of initiating warfarin in previously nonexposed valves and avoid potential confounding. A goal for future studies includes evaluating AVICs from patients receiving warfarin to assess if these cells demonstrate higher baseline activation of procalcific pathways compared with patients not taking warfarin. In addition future research may expand to include other valve types to determine if warfarin’s mechanistic effects translate globally to heart valves in other locations with exposure to different levels of mechanical stress and varying flow dynamics.
Other investigators have explored warfarin’s impact on vascular calcification. At therapeutic doses warfarin has been found to induce calcification in vascular smooth muscle cells.9 Warfarin has also been found to induce calcification in arteries and heart valves of rats.4 Furthermore clinical studies have identified an association between warfarin use and valvular calcification, demonstrating a significant link between warfarin therapy and risk of calcification as well as progression of aortic stenosis severity based on hemodynamic and anatomic measurements.6 However the effects of warfarin on the biology of the aortic valve have remained obscure. Additional prior work identified the ERK1/2 pathway as a mediator of osteogenic protein expression in AVICs. Thus it is no surprise that this present study confirmed that the ERK1/2 signaling pathway plays a role in warfarin-induced osteogenic protein expression.12 A significant body of work has highlighted the pathogenic role of Wnt/β-catenin signaling in activation of AVICs.13–15 However the mechanistic role of Wnt/β-catenin signaling in cellular osteogenic activity in AVICs has been unclear until present.
This present study elucidates a mechanism for warfarin-induced valvular calcification mediated by AVICs through ERK1/2 and Wnt/β-catenin signaling pathways. A simplified mechanistic summary for warfarin-induced osteogenic activation of AVICs is illustrated in Figure 10. Because warfarin is a commonly prescribed medication, it is imperative that we better understand the cardiovascular side effects of this medication. The association between warfarin and calcification of the aortic valve should be evaluated with large and robust clinical studies to confirm in vitro findings. Interestingly this present study also identified that upregulated Wnt/β-catenin signaling in diseased AVICs mediated cellular osteogenic activity induced by warfarin. Although these mechanistic insights inform our understanding of the osteogenic response to warfarin, at this time it is unclear whether this difference in signaling between normal and diseased AVICs is an acquired pathologic change in gene expression or an underlying genetic abnormality that was present in these patients from birth. Such analysis was beyond the scope of the present study, and further mechanistic work is needed to answer this question. Additionally the upstream effects of warfarin on the initiation of these pathways remain undefined and pose an important question for future work. In our experiments using warfarin plus dkk1 we show that inhibition of the extracellular receptor LRP6 by dkk1 inhibits warfarin’s effects. This indicates that warfarin, either directly or indirectly, acts on the LRP6 extracellular receptor. However it remains to be seen whether these effects are achieved through direct or indirect stimulation of the LRP6 receptor and whether upregulation in LRP6 activity is mediated by intracellular processes, extracellular processes, or both. Although warfarin’s ability to diffuse across cellular membranes and act directly within the cell is well established through its anticoagulant effects, this mechanism may not be universally used in all tissues and may differ specifically in regard to its procalcific effects. Exploring the upstream mechanisms responsible for warfarin’s effects in AVICs is an important endeavor for future studies. Overall the findings of this present study contribute to our understanding of warfarin-induced osteogenic activity in AVICs and may eventually lead to development of medical therapies for prevention or inspire change in clinical practice regarding choice of anticoagulant agents.
FIGURE 10.
Proposed mechanism of warfarin-induced osteogenic protein expression in aortic valve interstitial cells (AVICs). Note that in diseased AVICs both extracellular signal-regulated kinase 1/2 (ERK1/2) and β-catenin signaling are involved. However in normal AVICs only ERK1/2 signaling is activated. Despite this difference warfarin stimulates the production of osteogenic proteins runt-related transcription factor-2 (Runx-2) and osterix (Osx) in both normal and diseased AVICs. We postulate that the additional activation of β-catenin signaling in diseased AVICs may help to explain why diseased cells display a greater increase in osteogenic protein expression relative to normal cells.
In summary results of this study implicate warfarin in calcification of the aortic valve mediated through ERK1/2 and Wnt/β-catenin signaling pathways. The findings of this mechanistic study contribute to our understanding of warfarin as an activator of osteogenic phenotypic changes in AVICs and, consequently, as a potential contributor to development of aortic stenosis in patients on warfarin therapy.
Acknowledgments
This work was supported by the National Institutes of Health (HL106582).
REFERENCES
- 1.Nkomo VT, Gardin JM, Skelton TN, Gottdiener JS, Scott CG, Enriquez-Sarano M. Burden of valvular heart diseases: a population-based study. Lancet. 2006;368:1005–1011. [DOI] [PubMed] [Google Scholar]
- 2.Liu AC, Joag VR, Gotlieb AI. The emerging role of valve interstitial cell phenotypes in regulating heart valve pathobiology. Am J Pathol. 2007;171: 1407–1418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Meng X, Ao L, Song Y, et al. Expression of functional Toll-like receptors 2 and 4 in human aortic valve interstitial cells: potential roles in aortic valve inflammation and stenosis. Am J Physiol Cell Physiol. 2008;294:C29–C35. [DOI] [PubMed] [Google Scholar]
- 4.Price PA, Faus SA, Williamson MK. Warfarin causes rapid calcification of the elastic lamellae in rat arteries and heart valves. Arterioscler Thromb Vasc Biol. 1998;18:1400–1407. [DOI] [PubMed] [Google Scholar]
- 5.Krüger T, Oelenberg S, Kaesler N, et al. Warfarin induces cardiovascular damage in mice. Arterioscler Thromb Vasc Biol. 2013;33:2618–2624. [DOI] [PubMed] [Google Scholar]
- 6.Lerner RG, Aronow WS, Sekhri A, et al. Warfarin use and the risk of valvular calcification. J Thromb Haemost. 2009;7:2023–2027. [DOI] [PubMed] [Google Scholar]
- 7.Kaufman DW, Kelly JP, Rosenberg L, Anderson TE, Mitchell AA. Recent patterns of medication use in the ambulatory adult population of the United States: the Slone survey. JAMA. 2002;287:337–344. [DOI] [PubMed] [Google Scholar]
- 8.Beazley KE, Deasey S, Lima F, Nurminskaya MV. Transglutaminase 2-mediated activation of β-catenin signaling has a critical role in warfarin-induced vascular calcification. Arterioscler Thromb Vasc Biol. 2012;32: 123–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Beazley KE, Eghtesad S, Nurminskaya MV. Quercetin attenuates warfarin-induced vascular calcification in vitro independently from matrix Gla protein. J Biol Chem. 2013;288:2632–2640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rajamannan NM, Nealis TB, Subramaniam M, et al. Calcified rheumatic valve neoangiogenesis is associated with vascular endothelial growth factor expression and osteoblast-like bone formation. Circulation. 2005;111:3296–3301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhan Q, Song R, Zeng Q, et al. Activation of TLR3 induces osteogenic responses in human aortic valve interstitial cells through the NF-κB and ERK1/2 pathways. Int J Biol Sci. 2015;11:482–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yang X, Meng X, Su X, et al. Bone morphogenic protein 2 induces Runx2 and osteopontin expression in human aortic valve interstitial cells: role of Smad1 and extracellular signal-regulated kinase 1/2. J Thorac Cardiovasc Surg. 2009;138:1008–1015. [DOI] [PubMed] [Google Scholar]
- 13.Chen JH, Chen WL, Sider KL, Yip CY, Simmons CA. β-Catenin mediates mechanically regulated, transforming growth factor-β1-induced myofibroblast differentiation of aortic valve interstitial cells. Arterioscler Thromb Vasc Biol. 2011;31:590–597. [DOI] [PubMed] [Google Scholar]
- 14.Xu S, Gotlieb AI. Wnt3a/β-catenin increases proliferation in heart valve interstitial cells. Cardiovasc Pathol. 2013;22:156–166. [DOI] [PubMed] [Google Scholar]
- 15.Xu S, Liu AC, Kim H, Gotlieb AI. Cell density regulates in vitro activation of heart valve interstitial cells. Cardiovasc Pathol. 2012;21:65–73. [DOI] [PubMed] [Google Scholar]