Clones of Interstitial Cells From Bovine Aortic Valve Exhibit Different Calcifying Potential When Exposed to Endotoxin and Phosphate
Objective— Our purpose was to study in vitro whether phenotypically-distinct interstitial cell clones from bovine aortic valve (BVIC) possess different calcifying potential in response to endotoxin (lipopolysaccharide [LPS]) and phosphate (Pi).
Methods and Results— Among various clones of BVIC obtained by limited dilution technique we selected 4 clones displaying different growth patterns and immunophenotypes. Uncloned and cloned cells were treated with combinations of LPS (100 ng/mL) and Pi (2.4 mmol/L). Uncloned BVIC showed increased alkaline phosphatase activity (ALP) after treatment with LPS, which resulted in calcification after addition of Pi. Among BVIC clones, only Clone 1 (fibroblast-like phenotype) showed a relevant increase in ALP after LPS treatment in parallel with prevention of smooth muscle (SM) α-actin accumulation. No effect was observed in clonal cells harboring a more stable SM cell-like profile (Clone 4). None of the isolated clones calcified but mineralization was induced in the presence of LPS plus Pi when Clone 1 was cocultured with Clone 4 or after seeding on type I collagen sponges.
Conclusion— Endotoxin and phosphate can act as valve calcification promoters by targeting specific fibroblast-like interstitial valve cells that possess a unique procalcific potential.
Severe aortic valve calcification is characterized by a poor prognosis, unless patients undergo valve replacement. The identification of promoters, cells, and molecular pathways involved in the calcific degeneration of the valves is of paramount importance for the development of medical therapies capable of slowing down disease progression. Cellular-driven processes are now considered crucial for calcium deposition in the aortic valve.1 Although dystrophic calcification represents the main pathological finding, aspects of mature lamellar bone and endochondral bone formation have been described in the calcified aortic valves.2 Moreover, analysis of calcified leaflets have shown the expression of bone-related proteins (such as osteopontin, osteocalcin, alkaline phosphatase activity, osteoprotegerin, RANKL, bone morphogenic proteins), chondro/osteoblast-specific transcription factors (such as Runx2/Cbfa1 and Sox9), and mediators involved in osteogenic programs (such as Wnt/Lrp5/β-catenin pathway).3–5 On the whole, these data suggest that processes resembling molecular patterns involved in bone remodeling can take place during aortic valve calcification. Inflammatory molecules, modified lipids, infectious agents, and circulating levels of calcium and phosphate have been proposed as potential promoters of valve calcification.1,6,7 However, little is known about the effects of these factors on the calcifying behavior of aortic valve interstitial cell (VIC) subpopulations. We and others have previously shown in the porcine and human system that VIC represent a phenotypically heterogeneous population of cells mainly composed by fibroblasts/myofibroblasts/smooth muscle (SM) cells.8 Whether these cell types are endowed with an inherent or inducible procalcific potential in response to pathogenic factors is not known.
Uremic patients often experience an acceleration of calcium deposition in the vasculature, including the aortic valve.7 Increased level of serum phosphate which typically occurs in these patients is now considered relevant for the progression of vascular calcification. Seminal work by Giachelli’s group demonstrated that phosphate could induce a phenotypic switch of SM cells toward an osteoblast/chondrocyte profile and promote calcium deposition in vitro.9 However, the effect of phosphate on the VIC calcification process has been poorly investigated.
Inflammatory processes have been linked to the calcific degeneration of the aortic valve leaflets. Activated immune cells are present in calcified valves, and some inflammatory cytokines (such tumor necrosis factor [TNF]-α and interleukin [IL]-1) have been shown to promote osteogenic conversion of VIC.1 Some recent studies suggest that infectious agents can contribute to calcific valve degeneration,6 and that also endotoxin could be included into the list of procalcific factors.10 In this context is not known, however, whether the different VIC subpopulations possess a different calcifying potential when challenged with endotoxin.
To this end, we tested the procalcific effects of endotoxin and elevated phosphate levels on uncloned and cloned aortic bovine VIC (BVIC).
Materials and Methods
A detailed description of the methods is available online as supplemental materials (http://atvb.ahajournals.org).
Primary aortic BVICs were obtained from explants of intact aortic bovine valve leaflets. Clonal cell populations were isolated using a limited dilution technique. Characterization of BVIC clones was performed by immunofluorescence assays on cytocentrifugates using the following antibodies: anti-SM α-actin (SMA; Sigma), anti-MyHC-apla1 (type A nonmuscle myosin, NMM), anti-SM22 (Abcam), anti-SM myosin (SMM), anti-CD29 (VMRD Inc), anti–von Willebrand Factor (vWF; Dako), anti-CD45 (Dako), antiosteopontin (OPN; Abcam), antiosteocalcin (OC; Abcam), anti-3G5 (ATCC). The same primary antibodies were used to carry out an immunocytochemical study of the intact bovine aortic leaflets. At confluence in 6-well plates both uncloned and cloned cells were treated with different combinations of LPS (100 ng/mL; E.coli, Sigma) and phosphate (Pi; 2.4 mmol/L final concentration).
Calcium content and ALP activity were quantified colorimetrically using the o-cresolphtalein complexone and p-nitrophenol production methods respectively (Chema Diagnostica).
Flow cytometry and western blotting were performed using standard protocols to evaluate the expression of OC and SMA, respectively.
Selected clonal cells were seeded in bovine porous microfibrillar type-I collagen sponges (Davol) and treated with different combination of Pi (2.4 mmol/L) and LPS (100 ng/mL). Calcification of the collagen scaffold was established by staining of cryosections with von Kossa and Alizarin red using standard protocols. On adjacent cryosections the presence of apoptosis was studied by using the ApopTag Plus Fluorescein In Situ Apoptosis Detection Kit (Chemicon).
Data are expressed as mean±SD. Statistical analysis was performed by using Student t test for the intragroup analysis and ANOVA followed by posthoc Fisher’s LSD test for the intergroup analysis. Significance was accepted at P<0.05.
Phenotypic Cell Profile of Intact Aortic Valve Leaflet
Immunocytochemical analysis of cryosections from intact aortic valve leaflets has confirmed the existence of a phenotypic heterogeneity among BVIC. In the ventricularis layer cells expressing SM lineage markers such as SMA, SM22, and SMM were more prevalent. These cells were negative for OC and some of them stained for OPN (Figure 1, panel B). Conversely, in the spongiosa/fibrosa layer the majority of cells stained positive for NMM but were negative for SMA and SMM. In these layers sparse cells were found positive for OC (Figure 1A). Von Kossa staining showed no calcium deposition in the leaflets (not shown). To sum up, in the ventricularis BVIC expressed a myofibroblast/SM cell phenotype whereas in the spongiosa/fibrosa most BVIC can be mainly identified as fibroblasts.
Isolation and Characterization of BVIC Clones
Uncloned BVIC growth in culture consistently showed heterogeneous morphologies. The majority of cells showed a cobblestone aspect, whereas a minority of scattered BVIC displayed a spindle-shaped arrangement. Using a cloning procedure, we isolated 40 clones of BVIC. For the phenotypic analysis of the cells we selected 4 clones, representative of the major cell morphologies found among the isolated clones, which displayed marked differences in growth pattern and spatial arrangement at confluence (Table). Clone 1 (high growth kinetic, cobblestone morphology) maintained a stable morphology over several passages and did not form nodules in long-term culture. Clone 4 (slow growth rate, spindle-shaped morphology) on reaching confluence easily aggregated forming ridges and rare nodules. As for Clone 1, Clone 4 kept a quite stable morphology and growth behavior after several passages (up to 13). Clones 2 and 3 showed an intermediate phenotype compared to Clones 1 and 4, did not maintain a stable morphology with subculturing and, after 8 to 9 passages, displayed some features of senescence (arrest of growth, vacuolization and acquisition of flattened and enlarged shape).
Studies on cytocentrifugates showed that all 4 clones were negative for endothelial (vWF) and the hematopoietic marker CD45 (Table). All of them stained positive for CD29 (mesenchymal cell surface marker) and NMM. Only rare cells from Clone 1 were positive for SMA and none of them stained for SMM. All the cells from Clone 4 expressed SMA and a high percentage were also positive for SMM (Table; supplemental Figure I). Clones 2 and 3 showed to be more heterogeneous than Clones 1 and 4 in terms of SMA expression and did not stain for SMM. None of the cells from Clone 1 was positive for OPN, whereas a faint staining was observed for OC. The latter was not expressed by Clone 4, which stained positive for OPN (Table; supplemental Figure I).
Endotoxin and Phosphate Effects on ALP Activity Expression and Calcium Deposition in Uncloned BVIC
We then investigated the effect of LPS (100 ng/mL) and Pi (2.4 mmol/L) in promoting ALP activity expression and calcium deposition in uncloned BVIC. Cells treated with LPS showed a progressive time-dependent increase in ALP activity compared to controls and cells treated with Pi alone (Figure 2A). Combined treatment for 9 days of BVIC with LPS plus Pi also induced an increase in ALP activity comparable to LPS alone. However, no further increase in ALP activity occurred after this time point in the cells treated with both LPS and Pi (Figure 2A). Effect of LPS on ALP activity is dose-dependent (supplemental Figure II). Interest-ingly, addition of Pi to the culture medium was able to promote matrix calcification only in cells treated with LPS. No calcium deposition was observed in the cells treated with Pi alone or LPS alone (Figure 2B). Calcification appeared in vitro as dark granules disperse in the matrix (Figure 2C) and could be observed from 12 days of treatments onward (Figure 2B). In both short- and long-term cultures we did not find the formation of mineralized nodules.
Selective Response of BVIC Clones to Endotoxin and Phosphate Treatment
We then asked whether the isolated BVIC clones 1 to 4 display a diverse response to LPS stimulation in terms of ALP activity expression. After 12 days of treatment with LPS (100 ng/mL) only cells from Clone 1 exhibited a relevant increase in ALP activity (Figure 3A). A modest augment in enzymatic activity was also detected in Clone 3 and Clone 4. None of the clones showed mineralization of the matrix after treatment with LPS and supplementing the culture medium with Pi (2.4 mmol/L; Figure 3C), even in long-term cultures. Further studies were performed only on Clone 1 and 4 that maintain stable growth pattern without evidence for cellular senescence. As far as the immunophenotypic profile in these clones is concerned, Clone 1 expressed low levels of SMA when tested in immunofluorescence assays (Table) but Western blotting tests revealed that this clone showed an increase in SMA expression with culturing (Figure 4A). However, the concomitant treatment of Clone 1 with LPS prevented the accumulation of SMA. Conversely, endotoxin treatment of Clone 4 was not able to exert such an inhibition (Figure 4A). Flow cytometry analyses confirmed that OC was expressed at a low level by Clone 1 (about 50% of the cells) and almost absent in Clone 4. Interestingly, treatment of Clone 1 cells with LPS for 12 days increased OC expression (about 80% of cells) whereas this treatment had no effect on Clone 4 (Figure 4B). Because Clone 1 expressed high level of ALP activity and increased expression of OC we investigated whether this clone could be considered as an osteoblast precursor and retaining the potential to differentiate in osteoblast-like cells following treatment with osteogenic medium. After 15 days of treatment of this clone with an osteogenic inducing medium (a combination of ascorbic acid, dexamethasone and β-glycerophosphate), no effect was observed on both ALP activity expression and matrix mineralization (supplemental Figure IIIA and B).
We then tested whether Clone 4 could influence the functional properties of Clone 1 via a coculture experiment. Hence, these 2 clones were admixed in a 1:1 cell ratio and the combined culture treated again with a combination of LPS and Pi. After 12 days of treatment, calcium deposition could be observed only in the coculture setting (supplemental Figure IVA). At the end of the treatment period, cells in cocultures showed a peculiar spatial arrangement characterized by some cells (resembling those of Clone 4) giving rise to ridges and rare nodules which surrounded “islets” of cells (resembling the cobble-stone morphology of Clone 1) still showing a monolayer placement. In this particular setting calcium deposition seemed to be localized to the “ridges” (supplemental Figure IVB).
Calcification of Collagen Scaffolds by BVIC Clones
Despite the increase of ALP activity and OC after treatment with LPS, cells of Clone 1 were not able to induce calcium deposition when grown on the plastic substrate. We hypothesized that the limited ability of this clone to promote calcium deposition could be related to its particular growth pattern, adhesion properties and, perhaps, reduced production of collagenous extracellular matrix. To test this hypothesis, cells from Clone 1 and 4 were seeded on bovine microfibrillar type-I collagen sponges and treated again for 12 days with Pi (2.4 mmol/L) or Pi plus LPS (100 ng/mL). Histological analysis of the collagen scaffolds showed that in the presence of Pi alone Clone 1 cells were able to promote matrix calcification. Supplementation with LPS induced an even higher mineralization of the scaffold compared to Pi alone. Modest calcium deposition in the collagen sponges seeded with Clone 4 was observed after treatment with Pi and LPS (Figure 5A and 5B). Collagen sponges without cells but treated for 12 days with LPS and Pi did not give calcification (not shown).
We also quantified apoptosis in the 2 clones cultured in the collagen sponges. A high degree of apoptosis was observed in the calcified collagen scaffolds seeded with Clone 1 cells and treated with the combination of LPS plus Pi. A lower level of apoptotic degeneration was observed in the same clonal cells treated with Pi alone. Apoptosis was almost absent in untreated Clone 1 cells and in all the sponges seeded with Clone 4 cells (supplemental Figure V).
In the present study, we have shown that BVIC derived from bovine aortic valve leaflets display different clonogenic features and susceptibility to in vitro calcification induced by endotoxin and phosphate treatment. The inherent phenotypic heterogeneity of primary cultures makes difficult to dissect the specific contribution of each cell subset, if any, to the calcification process. Thus, cloning is the procedure of choice to study the specific structural and functional properties of cells within a giving tissue and investigate their potential role in some pathological conditions.
The existence of cells with a calcifying potential in the cardiac valves has already been demonstrated in vitro.11,12 Using an uncloned population of human and canine VIC, Mohler et al11 reported on the formation of calcified nodules after 3 to 4 weeks of culture. The nodules contained a central core of calcium, stained for BMP-2 and expressed ALP.11 In our study we expanded this original observation taking into account that BVIC, similarly to other nonbovine VIC,8 is in vivo comprised of a heterogeneous cell population as identified by an appropriate immunophenotypic score (Figure 1). In concordance with data reported for other species,8 bovine shows in their aortic valve leaflets a heterogeneous population of fibroblasts/myofibroblasts/SM cells specifically distributed among the 3 valve layers. In vitro treatment of uncloned population of BVIC with endotoxin and Pi was able to induce the expression of ALP activity and promote calcium deposition. Isolation of BVIC clones has targeted clones with a differential propensity for an inducible calcification pattern and this has been attained via endotoxin treatment. In particular, this promoter increases ALP activity and OC expression in Clone 1 (fibroblast-like phenotype). Thus, it is likely that in the uncloned BVIC population cells with this morphological and immunophenotypic profile play a major role in augmenting ALP activity. On the other hand, BVIC clones showing expression of markers of SM cell lineage (such as SMA, SM22, and SMM) appear to have a reduced ability of upregulating ALP and OC after stimulation with endotoxin. Interestingly, in vivo BVIC with an immunophenotypic pattern similar to Clone 1 are mainly localized to the spongiosa/fibrosa layer, whereas BVIC similar to Clone 4 (SM-like phenotype) predominate in the ventricularis layer. Although an in vivo direct spatial correlation with in vitro cell phenotypes cannot, at the present, be drawn, is a matter of fact that these layers are characterized by a different propensity to undergo calcific degeneration, as also demonstrated by a differential expression of calcification-related genes between the endothelial cells lining the two sides of valve leaflet.13
Current theories on the cellular involvement in aortic valve calcification include: (1) osteogenic differentiation of mesenchymal progenitor cells resident in the valve leaflet or arising from the bloodstream, (2) phenotypic transition of resident VIC toward a chondro/osteogenic phenotype, and (3) apoptotic degeneration of cells resident/recruited in the pathological tissue. The relative contribution of these mechanisms in promoting valve calcific degeneration is presently unknown. It must be pointed out however that: (1) our clones were not grown using culture procedures for mesenchymal stem cells, (2) bovine mesenchymal stem cells cultured with the BVIC medium containing LPS showed no ALP activity (unpublished data) and (3) Clone 1, the one that exhibits the higher plasticity in term of SM (SMA) and osteoblast (ALP, OC) markers expression, do not acquire the “fully” expression of an osteoblast profile when treated with a specific osteogenic differentiation medium. Although ALP is considered a nonspecific osteoblast marker, OC is presently used as indicator of osteogenic differentiation. The specific role of OC during bone mineralization is not defined yet and there are some evidence for the involvement of this protein in glucose and fat metabolism.14 Several histopathologic studies conducted in humans and in animal models of valve calcification show OC expression inside the pathological leaflets.3,4 In this context, the origin of OC-positive cells is not known. Our data suggest that a subpopulation of cells harboring a fibroblast phenotype (Clone 1) can indeed express OC and its expression increases under pathological stimuli. However, the lack of ability of these cells to spontaneously calcify in vitro or on treatment with specific osteogenic differentiation media suggests that they cannot be considered as mesenchymal stem cells.
Among the major determinants of extracellular matrix mineralization identified by Murshed et al15 are the following: phosphate levels, ALP expression, and the presence of type-I collagen. ALP and type-I collagen can be found in nonbone tissues but they are coexpressed only by osteoblasts in the bone and odontoblasts in the teeth. We identified Clone 1 as the cell type expressing the higher level of ALP after treatment with endotoxin but this response, though necessary, is not sufficient to promote calcium deposition in plastics by increasing Pi levels. However, when these cells are seeded on type-I collagen scaffolds, induction of ALP activity and high Pi level induced calcification of matrix. This finding suggests that Clone 1 is characterized by a limited ability to produce, at least when grown on plastics, the collagenous matrix needed for calcium deposition. Besides confirming the role played by these factors in driving the calcium deposition, our data demonstrate that a specific valve cell type reveals a unique procalcific profile under pathological conditions.
Nodule formation is often seen in cell culture models used to study the mechanisms of vascular calcification. Canine, sheep, and human aortic VIC cultures yield in vitro nodules, as well.11,16 However calcification of these VIC has not been studied using clonal cell populations. A clonal study was instead performed by Demer’s group to identify the presence in the bovine arterial wall of calcifying valve cells (CVC), a specific subset of SMC that spontaneously calcifies in vitro, forms nodules and retains a mesenchymal multilineage potential (chondrogenic, leiomyogenic, stromogenic).17,18 CVC are characterized by the expression of 3G5, a surface ganglioside that is also express by pericytes.19 Differently from CVC, Clone 1 cells even if express high level of ALP activity, are negative for 3G5 and do not form nodules even in a long-term culture. Interestingly, we observed that when Clone 4, characterized by low level of ALP but ability to form some nodules, was added to Clone 1, cells that morphologically resemble Clone 1 surrounded cells that grow in multilayers or nodules. In this setting, calcium deposition in the nodule region is favored (supplemental Figure IV). We speculate that the multilayered growth pattern of Clone 4 can favor the deposition of collagenous matrix used by the Clone 1 cells to complete the mineralization process (as seen when cells are seeded on the collagen sponges) via a process of “reciprocal functional cooperation.” However, the coculture data do not allow to draw a firm conclusion on the in vivo pathophysiological relevance of Clone 1–Clone 4 interaction inside the calcifying valve. Nevertheless, we cannot exclude that during pathological fibroproliferative remodeling of the leaflet, cells types previously located in separate areas can establish a closer contact and, hence, influence each other reproducing to some extent the coculture conditions.
The possible relevance of Pi during valve calcification is highlighted by previous in vitro studies that used phosphate donors (such as ß-glycerolphosphate) to promote calcium deposition by VIC.12 In our model, Pi alone was not able to promote calcium deposition when the cells were grown on plastics. However, when Clone 1 was seeded on collagen scaffolds the treatment with Pi was sufficient to induce the calcification of the matrix, accompanied by apoptosis of clonal cells. Previous studies carried out in both VIC and SMC have demonstrated that blocking apoptosis pathways can abate the in vitro process of calcification,16,20 suggesting that apoptosis could represent the final differentiation stage of cells acquiring a procalcific profile (as it happens at the end of chondrocyte maturation). Apoptosis cannot be considered per se a trigger for calcific degeneration of vascular tissues as suggested by Clarke et al,21 who showed that the chronic induction of apoptosis in vascular SMC was not accompanied in vivo by calcium deposition in the arterial wall. Nevertheless, increase in calcium deposition can be observed when apoptosis is induced in the context of atherosclerotic lesions, suggesting that the microenvironment and cell phenotypic modifications are the major determinants for the induction of calcium deposition. The interplay among phosphate, activation of apoptotic pathways, and matrix mineralization in our valve cells certainly deserves a further study.
Some studies hypothesize that microorganisms could promote calcific valve degeneration by inducing recurrent/low-grade endocarditis.6 Our data showing that LPS favor the acquisition of a procalcific profile by BVIC is in line with this hypothesis and open a new scenario on the role of microorganisms as possible players of valve calcification.
In conclusion, we observed that BVIC clonal subpopulations exhibit different calcifying potential in response to pathogenic factors. Specifically, we identified a peculiar subpopulation of BVIC characterized by a fibroblast-like phenotype that acquire a procalcific profile and promote collagen-matrix calcification in response to endotoxin and elevated phosphate levels.
We thank Dr Diego Guidolin for expert technical assistance.
Sources of Funding
This work was supported by FORIBICA.
Original received July 18, 2008; final version accepted September 21, 2008.
O'Brien KD. Pathogenesis of calcific aortic valve disease: a disease process comes of age (and a good deal more). Arterioscler Thromb Vasc Biol. 2006; 26: 1721–1728.
Mohler ER III, Gannon F, Reynolds C, Zimmerman R, Keane MG, Kaplan FS. Bone formation and inflammation in cardiac valves. Circulation. 2001; 103: 1522–1528.
Srivatsa SS, Harrity PJ, Maercklein PB, Kleppe L, Veinot J, Edwards WD, Johnson CM, Fitzpatrick LA. Increased cellular expression of matrix proteins that regulate mineralization is associated with calcification of native human and porcine xenograft bioprosthetic heart valves. J Clin Invest. 1997; 99: 996–1009.
Shetty R, Pepin A, Charest A, Perron J, Doyle D, Voisine P, Dagenais F, Pibarot P, Mathieu P. Expression of bone-regulatory proteins in human valve allografts. Heart. 2006; 92: 1303–1308.
Caira FC, Stock SR, Gleason TG, McGee EC, Huang J, Bonow RO, Spelsberg TC, McCarthy PM, Rahimtoola SH, Rajamannan NM. Human degenerative valve disease is associated with up-regulation of low-density lipoprotein receptor-related protein 5 receptor-mediated bone formation. J Am Coll Cardiol. 2006; 47: 1707–1712.
Ribeiro S, Ramos A, Brandao A, Rebelo JR, Guerra A, Resina C, Vila-Lobos A, Carvalho F, Remedio F, Ribeiro F. Cardiac valve calcification in haemodialysis patients: role of calcium-phosphate metabolism. Nephrol Dial Transplant. 1998; 13: 2037–2040.
Bertipaglia B, Ortolani F, Petrelli L, Gerosa G, Spina M, Pauletto P, Casarotto D, Marchini M, Sartore S. Cell characterization of porcine aortic valve and decellularized leaflets repopulated with aortic valve interstitial cells: the VESALIO Project (Vitalitate Exornatum Succedaneum Aorticum Labore Ingenioso Obtenibitur). Ann Thorac Surg. 2003; 75: 1274–1282.
Meng X, Ao L, Song Y, Babu A, Yang X, Wang M, Weyant MJ, Dinarello CL, Cleveland JC Jr, Fullerton DA. 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. 2007.
Kaden JJ, Bickelhaupt S, Grobholz R, Haase KK, Sarikoc A, Kilic R, Brueckmann M, Lang S, Zahn I, Vahl C, Hagl S, Dempfle CE, Borggrefe M. Receptor activator of nuclear factor kappaB ligand and osteoprotegerin regulate aortic valve calcification. J Mol Cell Cardiol. 2004; 36: 57–66.
Simmons CA, Grant GR, Manduchi E, Davies PF. Spatial heterogeneity of endothelial phenotypes correlates with side-specific vulnerability to calcification in normal porcine aortic valves. Circ Res. 2005; 96: 792–799.
Murshed M, Harmey D, Millan JL, McKee MD, Karsenty G. Unique coexpression in osteoblasts of broadly expressed genes accounts for the spatial restriction of ECM mineralization to bone. Genes Dev. 2005; 19: 1093–1104.
Clark-Greuel JN, Connolly JM, Sorichillo E, Narula NR, Rapoport HS, Mohler ER III, Gorman JH III, Gorman RC, Levy RJ. Transforming growth factor-beta1 mechanisms in aortic valve calcification: increased alkaline phosphatase and related events. Ann Thorac Surg. 2007; 83: 946–953.
Tintut Y, Alfonso Z, Saini T, Radcliff K, Watson K, Bostrom K, Demer LL. Multilineage potential of cells from the artery wall. Circulation. 2003; 108: 2505–2510.
Son BK, Kozaki K, Iijima K, Eto M, Kojima T, Ota H, Senda Y, Maemura K, Nakano T, Akishita M, Ouchi Y. Statins protect human aortic smooth muscle cells from inorganic phosphate-induced calcification by restoring Gas6-Axl survival pathway. Circ Res. 2006; 98: 1024–1031.
Clarke MC, Littlewood TD, Figg N, Maguire JJ, Davenport AP, Goddard M, Bennett MR. Chronic apoptosis of vascular smooth muscle cells accelerates atherosclerosis and promotes calcification and medial degeneration. Circ Res. 2008; 102: 1529–1538.